chelating carbene ligands and their metal complexeschelating carbene ligands and their metal...

Chelating Carbene Ligands And Their

Metal Complexes

By

Stuart Niven B.Sc. (Hons)

This thesis is presented for the degree of Doctor of Philosophy

to The University of Wales Cardiff,

Department of Chemistry

2006

The work described in this thesis was carried out in the Department of Chemistry

at The University of Wales Cardiff under the supervision of Prof. Kingsley J. Cavell and Dr

Ian A. Fallis

UMI Number: U585017

All rights reserved

INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.

In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed,

a note will indicate the deletion.

Dissertation Publishing

UMI U585017Published by ProQuest LLC 2013. Copyright in the Dissertation held by the Author.

Microform Edition © ProQuest LLC.All rights reserved. This work is protected against

unauthorized copying under Title 17, United States Code.

ProQuest LLC 789 East Eisenhower Parkway

P.O. Box 1346 Ann Arbor, Ml 48106-1346

DECLARATION

This work has not previously been accepted in substance for any degree and is not being concurrently submitted in candidature for any degree.

S ig n ed .^^^* ^T > ^^ t-rr.......................... (Candidate)

Date. ......................................

STATEMENT 1

This thesis is the result of my own investigations, except where otherwise stated. Other sources are acknowledged by footnotes giving explicit references. A bibliography is appended.

(Candidate)Signed

Date.. 2 l b / c o .

STATEMENT 2

I hereby give consent for my thesis, if accepted, to be available for photocopying and for inter-library loan, and for the title and summary to be made available to outside organisations.

Signed... c S — . ; . - . . - ................ (Candidate)

Date 2 j. Jo TTp?,...............................

Abstract

This thesis describes the synthesis of a number of functionalised imidazolium salts as precursors to N- heterocyclic carbenes and their subsequent coordination to Ag and Pd. Further a number of the Pd complexes were tested in the Heck reaction and their activities compared to complexes with similar structural features currently within the literature.

A range of imidazolium salts have been synthesised which include quinoline and octahydroacridine moieties and have been characterised by a number of methods including X-ray crystallography. A bis imidazolium salt has also been prepared as a DIOP analogue. The imidazolium salts were successfully reacted with Ag20 to form the NHCAg(I) complexes. The quinoline and octahydroacridine based NHCs were transmetallated to Pd as chelating ligands, the quinoline based systems appearing as planar, strained complexes in the X-ray structure.

The activities of the quinoline and octahydroacridine based NHCPd(II) complexes in the Heck coupling of 4-bromoacetophenone and 4-chlorobenzaldehyde with n-butyl acrylate were assessed and found to be comparable to similar systems with low to satisfactory conversions.

Chapter One - Introduction1.1 Classification Of Carbenes 1

1.2 A Brief History Of Carbenes 3

1.3 N-Heterocyclic Carbenes As Ligands 7

1.4 Chelating And Mixed Donor Ligands 8

1.5 Chirality In Carbenes 16

1.6 Routes To Forming Carbene Complexes 21

1.7 Catalysis And NHC’s 25

1.7.1 C-C And C-N Coupling Reactions 26

1.7.2 Mechanistic Aspects Of The Palladium Catalysed Heck Reaction 28

1.7.3 NHC’s As Efficient Ligands For The Heck Reaction 29

1.8 Aims Of This Thesis 31

1.9. References 32

Chapter Two-Ligand Design: Chelating Imidazolium Salts

And Their Synthesis2.1 Introduction. 41

2.2 Results And Discussion 44

2.2.1 Preparation Of Functionalized Imidazolium Salts. 44

2.2.1.1 Quinoline Based Imidazolium Salts. 44

2.2.1.2 An Octahydroacridine Based Imidazolium Salt 51

2.2.2 Bis Imidazolium Salts. 58

2.2.2.1 Preparation Of An Imidazolium Salt DIOP Analogue. 60

2.3 Conclusions 63

2.4. Experimental 63

2.4.1 General Comments 63

2.4.1.1 Synthesis Of Functionalised Imidazolium Salts 64

2.4.1.2 Synthesis Of Bis Imidazolium Salts 71

2.5. References 72

Chapter Three

Silver (I) Complexes Of Functionalized And Bis- N-Heterocyclic

Carbene Ligands

3.1 Introduction. 73

3.1.1 Ag(I)(NHC) As A Transmetallation Agent 73

3.1*2 Structural Characteristics Of Ag(I) Carbene Complexes. 78

3.2. Results And Discussion. 79

3.2.1 Ag(I) Complexes Of Quinoline Functionalised Carbenes 79

3.2.2 Ag(I) Complex Of Octahydroacridine Functionalised

Imidazolium Salt 81

3.2.3 Ag(I) Complex Of Bis-imidazolium Salt DIOP Analogue 83

3.3 Conclusions 86

3.4. Experimental 86


3.4.2 Synthesis Of Functionalised Ag(I) NHC’s 87

3.5 References 91

Chapter Four

Palladium Carbene Complexes

4.1 Introduction NHC Metal Complexes 92


4.2.1 Quinoline Functionalised Carbene Complexes Of Pd 94

4.2.2 Pd Complex Of Octahydroacridine Functionalised Carbene 98

4.2.3 Electrospray Mass Ionisation (ESI) Spectroscopy of 4.14 & 4.20 102

4.2.4 Pd Complex Of Bis Carbene DIOP Analogue 104

4.3 Conclusions 105

4.4 Experimental 106


4.4.2 Synthesis Of Functionalised Pd(II) NHC’s 107

4.5 References 111

Chapter Five

Pd catalysed C-C Coupling reactions : The Heck Reaction

5.1 Introduction 113


5.3 Conclusions 120

5.4 Experimental


5.4.2 Analysis Method 121

5.4.3 Procedure For The Heck Reaction 121

5.5 References 122

APPENDIXX-Ray Crystallography Data 123

Glossary

Ad Adamantyl

Ar Aryl

BA Butyl acrylate

BAP 4-Bromoacetophenone

BArF B[3,5-(CF3)2C6H3]4-

COD 1 ,5-cyclooctadiene

dba Dibenzylidineacetone

DCM Dichloromethane

DIPP 2,6-bis(diisopropyl)phenyl

DMAc N,N-Dimethylacetamide

DMF N,N-dimethylformamide

DMSO Dimethylsulfoxide

Et20 Diethyl ether

GC Gas chromatography

L Neutral, 2 electron donor ligand, e.g. phosphine or

carbene

LDA Lithium Diisopropylamine

M Metal

Me Methyl

Mes Mesityl or 2,4,6-trimethylphenyl

NHC N-heterocyclic carbene

OAc Acetate anion

Ph Phenyl

r.t. Room temperature

THF Tetrahydrofuran

X Anionic Ligand e.g. Halogen

Acknowledgements

I am indebted to Professor Kingsley Cavell and Dr Ian Fallis for taking me on as their

student and also for their supervision and support throughout the course of this work, both

during my time in the laboratory and the assistance provided whilst writing up from a

distance from Cardiff.

My appreciation is also extended to all the technical staff at Cardiff University, in particular

Alun, Rob, Gary and Robin for all the help provided when required as well as Li-ling for

the X- ray structure determinations.

Thanks must also go to all those members, past and present of Lab 2.84, who have been

inspirational as well as amusing! Thanks Dave, Dennis, Rhian, Nick, Vanessa, Huw, Emyr,

Maira, Mandeep, Gareth, Debs, Adrian and Emma.

I must also thank my Mum and Dad and Thuy for supporting me throughout my Ph.D as

well as ‘ensuring’ that I wrote up whilst working!

Chapter 1: Introduction

Chapter One

1.1 Classification Of Carbenes

A carbene is a neutral compound featuring a divalent carbon atom with only six

electrons in its valence shell1. Because carbenes do not posses an octet of electrons

they are generally highly reactive species. For the majority of carbenes the carbon

atom’s orbitals are bent to some degree away from linearity, which has an effect of

breaking the degeneracy of the nonbonding orbitals and causing the carbon to become

sp2 hybridized. The py orbital remains almost unchanged and is termed pn whilst the

px orbital is stabilized due to the acquisition of some s character and is called a

(Figure 1.1(a)). This break from degeneracy leads to different electronic

configurations which divide carbenes into two types - singlets and triplets. To form

the triplet state the two nonbonding electrons must occupy two different orbitals with

parallel spins, while in the singlet state they pair to occupy either the a or p% orbital.

The ground state spin multiplicity is a fundamental feature of carbenes that dictates

their reactivity2. A singlet state is favoured by a large o-pn separation and according to

Hoffmann, a value of at least 2 eV is needed to impose this. Anything below 1.5 eV'X - -ileads to the triplet state . The factors that cause the electrons to adopt one of these

configurations will now be considered.

The substituents on the carbene can determine whether a singlet or triplet state is

formed through a combination of inductive, mesomeric and steric effects, a- electron

withdrawing groups increase the o-pn gap by inductively stabilising the a nonbonding

orbital by increasing its s character and thus favouring the singlet state. Conversely, a

a electron donating substituent will induce only a small o-pn gap and favour the triplet

state. Mesomeric effects influence the ground state multiplicity to a much greater

extent and are the dominating feature. The mesomeric effect consists of the interaction

of the carbon pn orbitals with the appropriate p o r n orbital of the carbene substituents.

For illustrative purposes substituents that are n electron donating can be termed X,

while n electron withdrawing groups can be termed Y (Figure 1.1(b)). XX carbenes

are bents singlets, while most of the YY carbenes are predicted to be linear singlet


carbenes (though truly linear carbenes are extreme cases, bulky substituents can

stabilise triplet carbenes by increasing the carbene bond angle and thus bringing the

carbene frontier orbitals closer to degeneracy).

pn

(a) (b)

Figure 1.1: Diagram illustrating the loss in degeneracy of the non-bonding orbitals of

a free NHC

N-Heterocyclic carbenes (NHCs) (Figure 1.2, (a)) are the subject of this work and are

firmly placed within the singlet state. The nitrogen atoms (an X substituent) adjacent

to the carbon are responsible for the stability of the NHC by a push pull effect (Figure

1.2 (b)). The nitrogen atoms posses a lone pair of electrons which can n donate to the

formally vacant pK carbon orbital thereby stabilising it. Being more electronegative

than the carbon, the nitrogens can inductively remove electron density from the

carbene centre which serves to stabilise the carbene lone pair. NHCs are therefore

neutral two electron donor nucleophiles. Other heterocyclic carbenes containing a

sulphur or oxygen atom are able to stabilise the carbene in a similar manner.

Figure 1.2: (a): A typical NHC, (b): The ‘Push-Pull’ effect of the adjacent nitrogen

atoms which stabilise NHCs. M represents mesomeric effects whilst I is inductive.


1.2 A Brief History of Carbenes

Around the time that Fischer first discovered carbene complexes (1964)4, extensive

studies were being carried out on saturated N-heterocyclic systems by Wanzlick and

co-workers. Their work focussed on dimers of electron rich tetraaminoethylenes to

which they proposed an equilibrium existed between the free carbene and the dimer

and was termed the ‘Wanzlick Equilibrium’5"8 (Figure 1.3).

Phi PhiiNV

> = <N1

/

NiPh Ph

Ph

N

N>Ph

Figure 1.3: The ‘Wanzlick equilibrium’. Wanzlick proposed that the dimer dissociated13readily into two diaminocarbene ‘halves’ .

Contrary to this hypothesis, Winberg’s9 investigations into the metathesis reaction

between cyclic and acyclic enetetramines showed that the corresponding cross

products did not occur. Lemal10 and co-workers also showed that mixed olefins from

different enetetramines only occurred in the presence of electrophiles. On the basis of

these findings and in the absence of NMR evidence to support Wanzlick, the

existence of free carbenes and the Wanzlick equilibrium was excluded11. Recently

however equilibrium mixtures between free carbenes and tetraaminoethylenes have• • 1 1 1 been observed for some benzimidazolin-2-ylidenes " which confirms the Wanzlick

equilibrium for these species.

It wasn’t until 1991 that the first free carbene was successfully isolated. Arduengo and

co-workers successfully deprotonated an imidazolium salt by reaction with NaH to

give the free carbene 1.1, (Scheme 1.1). When stored under an inert atmosphere 1.1

was found to be an indefinitively stable colourless crystalline solid and bore an

unexpectedly high thermal stability; (melts at 240 °C without decomposition)14.


Initially the stability of Arduengo’s free carbene was thought to be due to the steric

bulk of the adamantyl groups preventing nucleophilic attack14. However isolation of

free carbenes with less sterically demanding groups such as 1,3,4,5

tetramethylimidazolin-2-ylidene15 and l,3-dimethylimidazolin-2-ylidene15 (Figure 1.4,

(1.2) & (1.3)) confirm that the stability arises from the mesomeric effects of the

carbene carbons adjacent nitrogens.

NaH

THF+ NaCI +

1.1

Scheme 1.1: Arduengo’s synthesis of the stable NHC l,3-bis(l-adamantyl)imidazolin-

2-ylidene (1.1).

Me Me

MeX>-N

N>:

Me

1.2 13

Figure 1.4: 1,3,4,5 tetramethylimidazolin-2-ylidene (1.2) and 1,3-

dimethylimidazolin-2-ylidene (1.3)

Prior to Arduengo’s discovery, little research was being carried out in the field of

carbenes due to the limitations of creating interesting novel metal complexes. The


method and isolation of 1.1 sparked a huge revival of interest in carbene chemistry

which is continuously growing.

Since the isolation of 1.1, many other stable cyclic (1.416,1.617,1.918, l . l l 19 and 1.1220)

and acyclic (1.519’21, 1.722,1.822 and 1.1023 ) carbenes have been synthesised which

include a mixture of heteroatoms, (Figure 1.5)24. Compound 1.11 is of interest as it is

the first example of a carbene to be stabilised by the donation of lonepairs from two

phosphines. (Note the presence of S and O in carbenes 1.6, 1.7 and 1.8 as being

sufficient substitutes for N). The isolation of 1.10 and 1.12 by Bertrand proved that

the presence of one nitrogen is sufficient to stabilise carbenes in certain cases. 1.10

was isolated as an indefinitely stable free carbene at room temperature and

demonstrated that providing a tertiary alkyl group is bound to the carbene carbon,

acyclic (alkyl)(amino) carbenes are isolable and importantly, behave as strong a

donors towards transition metal centres . 1.12, termed a ‘CAAC’ (Cyclic alkyl

(amino) carbene) was recently reported by the same group and is the first example of

a cyclic carbene where one of the electronegative amino substituents of an NHC is

replaced by a strong a donor alkyl group. This makes the ligand more electronegative

than diamino NHC’s. The quaternary carbon a to the carbene carbon atom also offers

a steric environment which is dramatically different from diamino carbenes and

phosphines. Further to the isolation of a number of CAAC’s, Bertrand successfully

prepared [PdCl(alkyl)(CAAC)] complexes as air stable crystals (stable enough to be

purified by column chromatography on silica gel) in high yields by the addition of

[{Pd(alkyl)(Cl)}2] to the corresponding carbenes. These were subsequently tested as

active catalysts in the Pd catalysed a arylation of ketones ' . The same group

recently reacted CAAC’s with CO to afford amino ketenes (Scheme 1.2 (a)) which are

indefinitely stable at room temperature. The formation of such species with diamino

NHC’s has been shown not to proceed by Arduengo (Scheme 1.2 (b)). The unique

steric and electronic properties of these CAAC’s coupled with Bertrand’s versatile

synthetic procedure, indicates that a wealth of CAAC ligands will no doubt follow

with potentially important applications.

Chapter 1: Introduction 6

0 >~ ' v N“ 0 >

( 1995)1.4

1.7(1998)

Y

1.5(1996)

1.8( 1998)

1.10(2004) 1.11

1.6(1997)

nN ^ N

( 1999)

1.9

(2005) (2005)

Figure 1.5: Examples of Isolated free carbenes with publication year.

•Pr

Dipp—NCO

'Pr

Dipp—N(a)

AdIN

CO

>: - *NI

Ad

AdI

N

NI

Ad

= C = 0 (b)

Scheme 1.2: The reaction of CAAC’s and traditional NHC’s with CO

1.12


1.3 N-Heterocyclic Carbenes As Ligands

Carbenes have been24 and still are compared to phosphines as ligands and it is easy to

understand why. Both ligands are neutral, strong two electron o donors, and have the

capacity to have their steric bulk manipulated. It has been shown however that an

NHC can form a much stronger bond with a metal than a phosphine31'33 and that n

back bonding is negligible. Studies have shown that even in the case of electron rich

late transition metals such as the linear Pd°(NHC) 2 or Pt°(NHC) 2 complexes, bond

formation is via donation of a electron density from the NHCs lone pairs into the

hybrid metal bonding orbital (dz2 + s)34. Backbonding from the filled metal d^ and dyx

orbitals into the carbene pn orbital is negligible, due to the occupancy of electron

density from the adjacent nitrogens . This absence of a requirement for backbonding

has meant that NHC’s have the advantage of being able to form stable metal

complexes with a range of metals that do not posses occupied orbitals and so would

not be able to participate in back-bonding. Metals such as Li ’ ’ , Be ’ and

hexavalent U39’40 have been shown to form stable metal complexes with NHCs. X-ray

diffraction data also confirms that bond lengths in NHC-M complexes are typical for

M-C single bonds1,33,65,112. The greater strength of the carbene bond means that in a

mixed complex of NHC and phosphine, the phosphine would dissociate in preference

to the carbene42. Using an NHC should therefore eliminate the need for the large

excesses of phosphine ligand in some catalytic reactions, where they are needed to

compensate for phosphine degradation by P-C bond cleavage.

The planarity of the heterocyclic ring, coupled with the space requirements of any N-

substituents means that NHC’s can be more sterically demanding than phosphines.

The majority of NHC complexes have also shown remarkable stability towards air

and moisture, as well as high temperatures. These advantages coupled with the

strength of the C-M bond and the ability to vary the N substituents, has given

carbenes an attraction for catalytic implications. When bound to metals, NHC’s are

significantly less reactive than the two major classes of carbene ligands, Schrock and

Fischer carbenes42, and have generally been viewed as spectator ligands, though

recent work has shown that they are not always such. It has been shown that NHC’s

can reductively eliminate to give 2-hydrocarbylimidazolium salts from hydrocarbyl-


M-carbene complexes43 8. Cavell first reported this decomposition pathway in 1998

during the examination of the synthesis and behaviour of complex 1.13 (Scheme 1.3).

Complexes with a Pd-Methyl bond in place of a halide (such as 1.13) are valuable as

pre-catalysts as they have been shown to display increased catalytic activity.

/

y H'

Me

N-Pd '

bf4-25 C Kl bf4-

COD Pd°N

1.13

Scheme 1.3: The reductive elimination of an imidazolium salt from a Pd hydrocarbyl

complex.

Complex 1.13 was found to undergo immediate reductive elimination at room

temperature generating the 1,2,3-trimethyl-imidazolium tetrafluoroborate salt as well

as free cyclooctadiene (COD) and Pd black49. Subsequent investigations have found

that bulky ligands accelerate this elimination reaction whilst electron rich ligands

decrease it. Bidentate ligands have also been shown to decrease the rate by restricting

the bite angle42. Consequentially this decomposition route via reductive elimination

has implications in catalytic reactions, particularly if intermediate complexes with

hydrocarbyl ligands are formed during the reaction.

1.4 Chelating And Mixed Donor Ligands

The major role played by bidentate phosphine and phosphite ligands in homogeneous

catalysis has naturally led to the development of chelating carbenes in an effort to

further exploit these strong a donor ligands. A wide range of ligands containing two

or more NHC groups are now known, a large number of these being modifications of

the general bis-carbene 1.14 (Figure 1.6). 1.15 is an example of a tri-NHC ligand

isolated by Dias and Jin in 199450. Further examples of bis carbenes are shown in the

section covering chiral NHC’s.


rrC(n)x

Kl/ ■ ' r ~ \ ’ \ /

/ V _N N-Buw

1.14 1.15

Figure 1.6: Examples of chelating carbenes.

Ligands containing both phosphorus and nitrogen donor atoms have been shown to

form strong metal phosphorus bonds and weak metal nitrogen bonds41 and are

important in many catalytic reactions51"55. Because of this many groups have made the

natural progression from unidentate mono carbenes towards mixed donor chelating

carbenes. The interest stems from the potential advantages a hemilabile ligand may

bring to a catalytic reaction. In general the hemilabile functionality means that whilst

the carbene is bound strongly to the metal centre, the hemilabile arm can offer both

stability to the metal via coordination, or allow vacant coordination sites by

dissociation. This could prove valuable in extending catalyst life in homogenous

reactions. A number of research groups have successfully incorporated a second

donor to the carbene ligand, with early progress being made with pyridine functions

as the N-substituents41,56,57. A number of examples are given in Figure 1.7.

X IN\ \\ 1.16 X 1 17

R-/( ^ V■Nkl/N

R= i-Pr Me3Si n-Bu Mes R

R= DiPP R= Dipp1.18 1.19Figure 1.7: Examples of pyridyl- functionalised NHC’s.

Mes1.20

1.16 was used to successfully form the PdMeCl(NHC) complex via transmetallation

from the Ag(I)NHC. Subsequent testing in the Heck reaction showed the catalyst to


have good stability and gave satisfactory conversions in the coupling of 4-

bromoacetophenone/4-chlorobenzaldehyde with n- butyl acrylate. High TONs were

also obtained in the Suzuki coupling of 4-bromoacetophenone with C6HsB(OH)2

f Z COthough at the expense of conversions . 1.16 was also used by Jin to form a cationic

chelating NHC Ni complex via the reaction of the Ag(I)NHC with Ni(PPh3)2Cl2 .

Preliminary studies showed the complex to be active in the polymerization of

norbomene and the polymerization of ethylene. Danopoulos also formed

PdMeCl(NHC) complexes from salts analogous to 1.16 where the second N

substituent is a lButyl or mesityl group. Both showed good activities for the Heck and71animation reactions . The potentially terdentate NHC precursor 1.17 was formed by

Chen and Lin41 and used to form Pd complexes in a monodentate and bidentate five

membered chelate fashion via the reaction of the salt with Pd(OAc)2 . Initial catalytic

testing of the copolymerisation of CO and norbomylene showed moderate to good

yields. 1.18 was used by Crabtree to form a Pd(0) complex via the oxidative

addition of an imidazolium C-H bond to the metal forming bis carbene complexes via

double oxidative addition. 1.19 is an example of a free heteroatom functionalised10'XNHC isolated by Danopoulos in 2002 . The corresponding imidazolium salt was

deprotonated using KN(SiMe3 )2 in THF between -10 and 0°C and crystallisation from

petroleum ether afforded the free NHC as air sensitive but stable at room temperature

crystals. The same group described 1.20 with a number of other functionalised NHCs

regarding their synthesis and structural studies .

Danopoulos states four reasons why a pyridine ring tethered to an NHC adds

versatility to the ligand design and they are, (i). The pyridine function is expected to

bind weakly to lower, softer oxidation states of the metal, (ii). This, in combination

with adjustment of the chelate ring size by using variable length linkers, can promote

hemilability with possible implications on the catalytic activity, (iii). The electronic

asymmetry of the chelating N-functionalised NHC ligand renders the corresponding

trans sites electronically inequivalent, due to the large difference in the trans effect of

the chelating ends. (iv). The donor and steric characteristics of the pyridine and NHC

functional groups are easily tuneable by a variety of substituents103.

Arnold reported the first amido functionalised NHC ligand and the synthesis of its

trivalent samarium(III) and yttrium(III) (1.21, Figure 1.8) adducts via the


i t

transamination of the corresponding Li carbene-amine . Other amido functionalised

NHC’s have been prepared by the group and used to form complexes with Uranium59

and Cerium60, while the group of Douthwaite61 prepared the first example of a

transition metal NHC-amide (1.22).

Imino functionalised carbenes have also been synthesised by various groups and

include 1.2362 and 1.2463 (Figure 1.8). Both were formed from transmetallation from

the Ag(I) carbenes. In 1.24 and its Rh(I) analogue the enamine tautomer forms upon

complexation with the Pd and Rh centres.

*Bu

N7 ^N(SiM e3)2\

Ln,N rN(SiMe.)3/2

Ln = Sm Y

N.

tBu1.21

r = \

N -Pd-C I

N N‘

tBu

tBu

w

1.22

Me/ = \

N. .N

Me—PdN i , Ar

1.23

/ = \Mes

Cl— Pd-N

1.24

Figure 1.8: Amido and Imino functionalised NHC complexes.

Alkoxide and phenoxide groups have also been used to functionalise NHC’s. Arnold

formed Li64, Cu64,65 and Ru66 (1.25, Figure 1.9) complexes of the former, while

Hoveyda67 and Grubbs68 formed Ru (1.26) and Pd (1.27) complexes of the latter

respectively.


i Mes—N

Ar

Me—P d -0

R

1.25 1.26 1.27

Figure 1.9: Alkoxide and Phenoxide functionalised NHC’s.

Because NHC’s and phosphines are both strong trans effect/influence ligands they are

both able to activate/labilise groups coordinated trans to themselves. In mixed NHC-

phosphine chelating square planar complexes the remaining groups attached to the

metal centre can evidently not avoid a position with a strong trans influence. This is

particularly important if the trans group is a migrating hydrocarbyl species in for

example the hydroformylation reaction where the migration process will be promoted.• • 6Q 7rt 22 onA number of examples of this class of ligand now exist ’ ’ ’ . Nolan was first to

show the high efficiency of a phosphine-imidazolium salt in the in-situ Heck coupling

of an aryl bromide with n-butyl acrylate69. Danopoulos published similar ligands in

complexes with Pd(II)70 as well as the free carbene103. The crystal structures of

Danopoulos’s complexes 1.28 and 1.29 showed interesting results. The two Pd-CH3

bond lengths of 1.28 were approximately equal while the Pd-Br bond in 1.29 trans to7 nthe phosphine was actually longer than that trans to the NHC . This is in contrast to

what would be expected due to the NHC’s stronger a donor strength and hence

greater trans influence.

r \A r N N Ar'

f = \NL .N

Me— Pd—PPh.I

Me

Br— Pd—PPh,IBr

1.28 1.29

Figure 1.20: Danopoulos’s NHC-phosphine complexes.

As previously stated chelating carbenes are also significantly less prone to reductive

elimination pathways than monodentate ones. This is possibly due to the NHC being


unable to adopt the correct conformation for reductive elimination due to the

with catalytically interesting Pd hydrocarbyl species and will be discussed later. It

should be noted that because of the rigid geometry of the five membered planer rings

of the NHC, the optimum number of atoms to form an appropriate chelate does not

match that for diphosphanes or diamines. Methylene bridged 2,3-dihydro-1H-

imidazol-2-ylidenes and l,2,4-lH-triazol-5-ylidenes form six membered rings upon

metal complexation. However the natural bite angle of 78-79° found in transition

metal complexes (W, Pd) of these ligands is exactly that of 1,1’-

bis(diphenylphosphino)methane (four membered ring) .

Tridentate ligands have also appeared in the literature as so called ‘pincer’ carbenes,

both with 2:1 and 1:2 carbene to mixed donor ratios. The pincer framework serves to

stabilise M-L bonds. Amongst those in the literature, Crabtree reported the first Pd

Pincer NHC complex (1.30)74 (Figure 1.21) as an essentially flat and rigid molecule

and as an active catalyst in the Heck reaction. Its low solubility in most non-polar• • 7 csolvents however has reduced its application in other Pd catalysed reactions and as a

remedial step Crabtree replaced the methyl wingtip groups for n-butyl to give

complex 1.31 which showed both improved solubility and catalytic activities. This

increase in solubility is thought to arise from the n-butyl groups causing a deviation

from planarity which has an effect of decreasing intermolecular stacking in the solid

state. The same group also later reported the Ru pincer complexes 1.32 and 1.33. 1.32

was reported to be active for hydrogen transfer as well as the oxidation of olefins

while 1.33 was shown to be completely inefficient at both reactions76.

conformational rigidity imposed by the chelate72. This is significant when dealing

+

Br

1.30 1.31


/ « ^ I \nBu Br q q nBu

, - 9N - ^ R u - N

(pf6)6/2

nBu nBu

1.32 1.33

Figure 1.21: Examples of ‘Pincer’ Pd complexes reported by Crabtree.

1 71Cavell synthesised the first Pd- hydrocarbyl pincer complex 1.34 (Figure 1.21),

which showed good stability with respect to reductive elimination up to 100°C in

DMSO and also proved to be an efficient catalyst in the Heck reaction. The analogous

complexes 1.35a and 1.35b containing the comparatively weaker amino donors were

also prepared. 1.35b was shown to give the reductive elimination product after only a

short period in solution while 1.35a was afforded as a stable solid. Cavell also77reported the Pd- hydrocarbyl complex 1.36 which was shown to have high stability

and good activity in the Heck reaction with activated aryl bromides.

r= \

Me

r y a\ —Pd /~N ^ N -

a = Br b = Me

1.34 1.35 1.36

Figure 1.21: Reported examples of Pincer complexes by Cavell.

Around the same time that Cavell published 1.36, Danopoulos synthesised the1 finanalogous complexes 1.37 and 1.38 and also demonstrated their stability and

activity in the Heck reaction. The same group also went on to synthesis 1.3978 as well

as reporting Co and Fe pincer complexes. The Fe complex 1.40 is the first to feature

C2 and C5 metallation modes coexisting on the same metal centre79. The same paper

also demonstrated the synthesis of the paramagnetic complex 1.41 by direct

metallation of the imidazolium salt with [Fe(N(SiMe3)2], the structure of which was

determined by X-ray crystallography. 1.41 was later used to synthesis the first

dinitrogen complex stabilised by NHC’s (1.42)80. This ‘aminolysis’ method of


carbene complex synthesis was further extended by the same group via reaction of

Co[N(SiMe3>2 ]2 with a bis imidazolium salt to give 1.43. 1.43 was subsequently used

to form a number of complexes including 1.44 and 1.45 by reaction of Na(Hg) ando i

MeLi respectively .

NI

Ar

Cl

1.37

Ar = MesAr = (2,6J Pr2C6H3)

N-I

Ar

IX

Ar

N ^ N

PdICl

1.38

N\Ar o - t - oN

IAr

Br

1.39

N-I

Ar

H Ar

V Ar

(FeCI4)

INBr'

Fe'Ar Br

Ar = (2I6JPr2C6H3)

N\Ar

y ~ ~ JAr ' N' 1Ar NN

N

N\Ar

1.40 1.41 1.42

X= Br, Me^isT Na(Hg) or MeLi

N Co" "" N / _ / \ \Ar Br Br Ar

1.43 1.44/1.45

Figure 1.22: Examples of Danopoulos’ pincer carbene complexes.

CoAr Ar

O ')

Gibson has reported a host of complexes according to Scheme 1.4 and tested these

in the oligomerisation and polymerisation of ethylene. Whilst the Fe complexes

showed evidence of alkyl carbene coupling, the V, Cr and Ti complexes were shown

to be less prone to this chemistry, with the Cr system showing particularly high

activities in the catalytic runs. These observations hint towards the usefulness of earlyA O 0 9

transition metal complexes of NHC’s in olefin oligomerization and polymerization ’ .


N N N 5 \ _ ,N— • \ W N

R RR = 'Propyl, L1 R = 2,6-diisopropylphenyl, L2 R = 1-adamantyl, L3

THF

Xn

TiCI3L2VCI3L2CrCI3L1CrCI3L2CrCI3L3CoCI2L2[FeL12][FeBr4]FeBr2L2FeCI3L2

Scheme 1.4: Gibson’s pincer carbene complexes82

The Pd PCP pincer complex 1.46 was prepared by Lee in 2004 and is an example of a

pincer containing three strong a-donors. This was shown to be effective in the Heck

and Suzuki reaction, although longer reaction times were needed with unactivated

substrates . As an indication of the potential versatility of NHC complexes, the Ag

NHC pincer motif 1.47 was functionalised with hydroxyl groups to give a water

soluble complex and was shown to be a useful antimicrobial agent84.

! = \ ISL .N

P P h r ^ P P h , Cl

N

N

<N

CH2)n

OH

N) “ Ag-

N

|CH2)n

OH

n = 2,3

1.46 1.47

Figure 1.23: Further examples of the pincer carbene motif.

1.5 Chirality in Carbenes

The addition of chiral centres to NHC’s is a logical development due to the obvious

use of chiral ligands in enantioselective catalysis. The number of chiral NHC’s

appearing in the literature has increased over the years with two reviews coveringO f 0 £

these up to 2004 ’ . Many of the early examples reflect familiar themes in

asymmetric synthesis: the use of pinene- and camphor- derived chiral auxiliaries,


oxazolines from optically active amino acids, atropisomerisms in binaphthyl systems

and the planar chirality of ferrocene85. The understanding of the key structural

elements which induce high stereoselectivity is still under development due to the

distinctly different stereochemical ‘topography’ of NHCs from diarylphosphines.

NHCs will not create an ‘edge to face’ arrangement of their aryl substituents - a

structural feature common to many chiral diphosphines such as derivatives of DIOP

and Binap etc.86,87 In their review86 Gade et al define six classes of ligand

characterised by the position of the chiral structural motif in relation to the donor unit.

1. NHC’s with N-substituents containing centres of chirality.

The ability to vary the N-substituents of NHCs means that the introduction of any

substituent containing a chiral centre is limited only within the constraints of the

synthetic methods of NHCs discussed in Chapter two. Herrmann88 and Enders89 first

formed chiral NHCs of this type in 1996 (1.48,1.49, Figure 1.24).

R r =f = \ /

I^ -R h -C I

R = Ph Nap

1.48

c i R-N

Y 'n R = CyN - ^

1.49

1.50 Ad

BArF-

1.51

Figure 1.24: Examples of NHC’s possessing chiral N-substituents.

1.48 showed good activity but low stereoselectivity for the hydrosilylation of

acetophenone (32% ee). 1.49 was generated as a mixture of diastereoisomers as a

consequence of the non symmetrical carbene ligand. The mixture was tested in the


hydrosilylation of methyl ketones giving low to moderate enantioselectivites (ee’s up

to 44%). Hartwig’s90 ligand 1.50, was tested as a stereodirecting ligand in the Pd

catalysed asymmetric oxindole reaction and gave superior results to those of

established chiral phosphines. Burgess91 successfully applied 1.51 in the

hydrogenation of unfunctionalised alkenes with an enantiomeric excess of >90%

which is comparable to the best phosphine or phosphate-oxazoline iridium

complexes92'94. Gade concludes that NHCs containing chiral N- substituents may be

efficient as stereo directing ligands if the N substituents are either very sterically

demanding or locked in fixed conformations, and that chiral induction is greater the

closer the chiral centre is located with respect to the N- substituents. This type of

chiral NHC generally gives moderate results in asymmetric catalysis.

2. NHC ligands with chiral elements within the ring.

The 4 and 5 positions of NHCs may be rendered chiral centres by an appropriate

substitution and the chiral information transmitted to the metal centres active site via

the N substituents. Systematic studies by the groups of Mangeney and Alexakis95,96

employed chiral imidazolinylidenes of this type in the alkylation of a enones. In their

studies the Cu NHC catalyst, generated by Ag(I) transmetallation, was used for the

asymmetric addition of diethyl zinc to cyclohexanone. One of the observations from

these studies showed that the two methyl N- substituents of 1.52 (Figure 1.25) were

inefficient in transmitting the chiral information from the C4 and C5 positions (ee

23%) whereas the steric repulsion between the t-butyl groups and the benzyl groups of

1.53 leads to a C2 symmetry arrangement with respect to the carbene donor function

thus transmitting the chirality to the reaction centre. 1.52 produced an ee of 58%.07 ORGrubbs ’ also synthesised chiral NHCs with N- aryl substituents of this type and

employed them in the stereoselective ring closing metathesis of olefins. Helmchen79

formed two Rh(I) diastereomeric atropisomers (from 1.54) of a chelating NHC-

phosphine ligand via transmetallation of the Ag(I) complex. These were subsequently

tested in the catalytic hydrogenation of dimethylitaconate and Z-acetamidoacrylate

with relatively high yields and excellent enantioselectivities (98-99%).


Agl

»Bu 'Bu

AgCI

1.52 1.53 1.54

Figure 1.25: Examples of chiral NHC’s with a chiral ring.

It is evident from these and other examples that chiral information from the NHC

backbone may be transmitted to the metal centre in varying degrees depending on the

nature of the N- substituents involved. The combination of chiral N substituents with

a chiral NHC ring is an area yet to be frilly explored.

3. NHC ligands containing an element of axial chirality.

Rajanbabu100 published the first chiral bidentate bis carbene complex of Pd (1.55,

Figure 1.26) based around the l,r-binaphthyl unit as the chiral element. The chirality

arises from the blocked rotation around the C-C axis linking the two naphthyl units

giving configurationally stable atropoisomers. On complexation with Pd the ligand

forms cis-trans mixtures as a result of the flexibility of the ligand and as such, no

stereoselective catalysis has been employed with this ligand. A related bis carbene

was reported by Min Shi101 and has the NHC’s directly linked to the binaphthyl

backbone. The Rh(III) complex 1.56 was synthesised and employed in the asymmetric

hydrosilylation of ketones, giving good activity and excellent enantioselectivity for

aryl alkyl ketones.

1.55 1.56

Figure 1.26: Examples of NHC complexes with axial chirality.


Hoveyda’s complex67 (1.26, Figure 1.9) and complexes of the groups subsequently

modified ligands, were also used in asymmetric olefin metathesis and showed

excellent enantioselectivities for asymmetric ring opening cross metathesis.

4. NHC’s containing an element of planar chirality.

1.57 (Figure 1.27) was the first reported planar chiral NHC by the group of Bolm in

2002, though its application in the Rh catalysed hydrosilylation of ketones yielded| ryy

only a racemic mixture of secondary alcohols . A number of other NHC ferrocence

derivatives have also appeared in the literature including those with mixed second

Figure 1.28) and Douthwaite104 (1.59) (1.60). 1.59 was tested in Pd catalysed allylic

donors and have shown a variety of activities in enantioselective catalysis.

SiMe3

1.57

Figure 1.27: An example of an NHC with planar chirality.

5. NHC’s joined by a chiral trans cyclohexadiamine ligand backbone.

Trans- cyclohexadiamine has been used as a chiral backbone by both Burgess103 (1.58

alkylations and led to high enantioselectivities, while 1.60 showed poor selectivity in

the enantioselective a- arylation of amides.

V ^ N

w

1.58 1.59 1.60

Figure 1.28: Examples of NHCs on a chiral backbone.


6. NHCs incorporating oxazoline units.

Herrmann reported the first bidentate chiral carbene containing an oxazoline ring

(1.61 Figure 1.29)105. The oxazoline unit has been established for a number of years in

asymmetric catalysis with its key features being its rigidity and quasi planarity.

Burgess’s catalyst106 (1.62) showed excellent activity and enantioselectivity for the

asymmetric hydrogenation of E-l,2-diphenylpropene, the author attributing the high

selectivity to the bulky 2 ,6 -(1Pr)2C6H3 group blocking one of the quadrants of the

active catalyst space, allowing control of the geometry of the coordination reaction

sphere.

+

BArF-

1.61 1.62

Figure 1.29: NHCs incorporating oxazoline units.

1.6 Routes To Forming Carbene Complexes

There exist a number of different routes to forming NHC(M) complexes. The most

appropriate route however, depends on the nature of the carbene precursor itself, as

well as the desired substituents on the metal centre.

The most obvious route is via the free carbene and is probably the best route for

forming Pd(0) NHCs and simple Pd(II)MeX(NHC)s. This is usually achieved by

deprotonation at the C2 position of imidazolium salts with a suitable base. KO'Bu

used in a stoichiometric amount or NaH/KH in the presence of a catalytic amount of

KO'Bu has been used in THF at room temperature14,73,107,108. In base sensitive

functionalised salts where the C2 proton may not be exclusively removed due to

acidic protons on linkers etc, an amide may be used and include Li(N'Pr2), (LDA) or

R = Me, !Bu R' = Bn, 'Pr

DIPP N

^ lr

Ad

N,

N-

o


K[N(SiMe3)2]108-112. If the free carbenes prove to be too unstable to be made at room

temperature, a liquid ammonia route may also be used where the azolium salt is

dissolved or suspended in a mixture of liquid ammonia and a polar aprotic solvent and

deprotonated at very low temperatures by NaH113,114 or BuLi115,119. Khun

demonstrated that free carbenes could also be generated by reacting imidazole-2 -

thiones with potassium in refluxing THF. This was achieved with 1,3,4,5-tetramethyl-

2-(methylthio)imidazolium iodide to give the product in good yield116. The free

carbene may also be formed via the thermal elimination of small molecules such as^ 1 Ovl 1 1 0*7 1 iC

MeOH from imidazolidines ’ ‘ , benzimidazolines and 1,2,4-triazolines

(Scheme 1.5) to yield the corresponding NHC. This method however is limited to

thermally stable NHCs. Once formed stoichiometric amounts of free NHC can

displace other donor ligands quantitatively from the coordination sphere of suitable

Pd(O) 3 4 1 1 7 ' 120 and Pd(II) 4 9 '72 108 110' 123 precursors. Displacing 1-5 cyclooctadiene (COD)

from Pd(II)XY(COD) precursors is particularly valuable when X is a methyl and Y az o n e \ <71

halide ’ ’ . The free carbene route suffers where the high air and moisture sensitivity

of the free carbene is concerned meaning generation and manipulation of these

species may be problematic.

PhPh ‘ I

OMe Heat, vacuumI1 X -----► T >:N -'N H - MeOH N - .* /

1 IPh p h

Scheme 1.5: The thermal elimination of MeOH.

Another route to the formation of metal complexes is by reaction of an imidazolium

salt with a metal containing precursor of sufficient basicity to deprotonate the organic7T . _ ,substrate . The formation of a strong metal-carbene bond is presumably an important

factor in driving the reaction. This was the method used by Ofele128 (equation 1,170 •Scheme 1.6 ) and Wanzlick (equation 2) who made chromium and mercury carbene

complexes respectively and this approach has been widely employed in the

preparation of many other NHC complexes. The application of Pd(OAc)2 in this

manner has led to a number of mono and bidentate imidazol-2 -ylidene and triazol-5 -

ylidene Pd(II) complexes being prepared113,130,131. Each of the basic ligands of the M


precursor deprotonates one imidazolium salt resulting in the new metal complex and

acetic acid73. The general formula of this Pd(OAc)2 route is Pd(NHC)2X2 . The use of

ramped temperatures and DMSO has also been developed which has shown an

improvement in yield for certain chelating Pd(II) dicarbene complexes132'135. The

Pd(OAc)2 route led to the first example of an NHC bound through the C5 carbon.

Reported by Nolan the reaction of N, N’ Bis(2,4,6-trimethylphenyl)imidazolium

chloride with Pd(OAc)2 in dioxane at 80°C led to the formation of 1.63 (Figure 1.30),

though reactions giving abnormal C-4 and C-5 metallation are very sensitive to

conditions as well as counter ions166. Surprisingly 1.63 also proved to be a better cross

coupling catalyst than certain complexes with conventionally bound carbenes136.

N[HCr(CO)s]-

Heat

-H,

r 'N[I ) — Cr(CO),

N(1)

6h5

NIc 6h5

CI04-Hg(OAc),

-2AcOH

<f6H5 <f6H5

(f y Hg— j )

c 6h5 c 6h5

2+

2 CIO. (2)

Scheme 1.6 : Reaction of imidazolium salts with basic metal precursors.

/ = \Mes^ Mes

Cl— Pd-CIMes

Mes

Cl

1.63

Figure 1.30: The first example of an NHC bound through the C5 carbon.


Oxidative addition of imidazolium or amidinium species to low valent metal centres

also yield M(NHC) species. In 2001 Cavell demonstrated the facile formation of NHC

complexes via the oxidative addition of 2-haloimidazolium salts to Pd° substrates137

1 W(Scheme 1.7, Equation 3). Furstner and co-workers also utilized the oxidative

addition of C2-X in their synthesis of metal carbene products, and the groups of

Nolan139 and Crabtree140 (Scheme 1.7, equation 4) provided evidence for the oxidative

addition of imidazolium salts to Pd141, although no Pd-hydride complex was isolated.

Oxidative addition to a C-H bond is a most desirable means for carbene complex

formation as no base is used, atom economy is maintained and the product can be the

potentially catalytically active species. After publishing density functional

calculations on the oxidative addition of 2-H imidazolium salts to zero valent group

10 metal precursors134 to form M(II)(Hydrido)(NHC) complexes and the initial

synthesis of Pt Hydrido(NHC) complexes, Cavell further supported this oxidative

addition route by forming Pt bis-(NHC) complexes via C-H activation at room

temperature as well as the synthesis of (NHC) M-Hydride complexes of Ni and Pd 138

(Scheme 1.7, Equation 5). Both the Ni and Pd complexes were shown to be

surprisingly stable both in solution and in the solid state.

N

H

Br

Pd-Br

(R = i-Pr, n-Bu)

(4)


Scheme 1.7: Oxidative addition for NHC complex formation.

In cases where the functionalised NHC precursor contains acidic protons about its

linker chain or those in which the free carbene is not easily accessible, an NHC

transfer reaction can be valuable. The use of Ag(I) NHC complexes has become

particularly useful as a transmetallation agent. The Ag(I) complex can be prepared via

the reaction of imidazolium salts with the weakly basic Ag2 0 , which is then reacted

with an appropriate metal precursor such as PdX2(COD) to give the PdX2(NHC)

complex. The silver transfer route is described in more detail in Chapter 3. The use of

other transition metal complexes have also been reported to transfer NHCs to Pd.

(NHC)M(CO)5, where M is W, Mo or Cr has been used in the transfer of saturated

NHC ligands onto a range of metals in good yield142,143.

The co-condensation of Ni, Pd and Pt vapour with l,3-di-N-tert-butylimidazol-2-

ylidene has been shown to produce stable, two coordinate homoleptic metal carbene

complexes of each respective metal. This method proved to not only be a straight

forward novel synthetic method, but also produced a complex which was previously

inaccessible by solution techniques144.

Complexes bearing saturated NHCs can be prepared by the thermal cleavage of

certain electron rich tetraaminoethylenes in the presence of appropriate metal

precursors. This method was used by Lappert and co-workers in the 1970s to form a10Trange of NHC-M complexes.


1.7 Catalysis and NHCs

A catalyst is a substance that brings about or accelerates a specific reaction or reaction

type, which proceeds in parallel with any other existing thermal or catalytic reaction

within the particular system145, but does not affect the position of the equilibrium.

Thus a catalyst affects the kinetics of a reaction and not the thermodynamics, and

although a catalyst participates in chemical reactions it is regenerated back to its

original state after it has completed its catalytic cycle (excluding catalyst poisoning or

degradation).

Specificity is a key factor in homogeneous catalysts. The market share of

homogeneous catalysis (1999) is low, around 10-15%148 and is a reflection of some of

the economic disadvantages of classical homogeneous catalysts. However the current

developments of transition metal complexes as catalysts has opened up the

possibilities of utilising these complexes in new C-C and C-N coupling processes

using simple substrates and reaction conditions which were not possible using

traditional synthetic methods.

1.7.1 C-C and C-N coupling reactions.

Carbon-carbon and carbon-nitrogen bond forming reactions are key steps in the

synthesis of many organic chemicals, with many of these cross coupling reactions

mediated by palladium and nickel catalysts. Catalysed cross coupling reactions such

as Kumada149'151, Negishi149’152, Stille149,153 and Suzuki149,154' 156 also make use of

transmetallating agents including organoboron, organomagnesium and organozinc

reagents. A summary of some of these Pd catalysed coupling reactions are given in

Figure 1.31. The widespread uptake of these Pd catalysed cross coupling reactions

have previously been impeded by the high cost of the aryl iodides and bromides

traditionally used in such reactions. Industry, and indeed academia, has looked

towards using aryl chlorides to replace these, due to their low cost and high stability.

This high stability however also prevents the aryl chlorides from readily undergoing

oxidative addition to 14 e- Pd(0) phosphine complexes157,158 and is the reason why it

has been hard to find general coupling protocols that are economically sound.


Heck

SonogashiraStille

ArSnBu

SuzukiArB(OH).

R'MgBr, R 1

H N vKumada R2R'ZnBr

Amines

NegishiR2

Figure 1.31: A selection of C-C and C-N catalytic reactions mediated by Pd. A

number of papers and reviews cover these reactions in more detail149,154’155,159"166.

In recent years research has shown that electron deficient substrates have efficientlyi f»nundergone the Heck reaction at room temperature and deactivated aryl chlorides

can also be coupled in Heck, Suzuki and aryl animation protocols through the

application of appropriate catalyst systems168,169. This has been largely brought about

by careful selection and modification of the electronic and steric properties of the

catalysts ligands. Coupling reactions are therefore suitable areas of application for

NHC complexes due to the relative ease of varying the ligands properties and the

current fertility in the area. A number of examples of carbene complexes utilised as

catalysts have already been given in other sections of this chapter. Due to the


relevance of the Heck reaction in this work the rest of this section will focus on this

well studied reaction.

1.7.2 Mechanistic Aspects Of The Palladium Catalysed Heck

Reaction

The generally accepted mechanism for the Heck reaction is given in Figure 1.32. The

reaction involves the cycling of a Pd centre between its 0 and 2+ oxidation

states88,117,130,170'172and it has been suggested that in particular systems when the Pd

complex contains a phosphinito PCP pincer ligand, the Pd centre cycles between the117 11'X 17^2+ and 4+ oxidation states ’ “ . Herrmann discussed the possible mechanisms

involved for palladacycles with Pd(0)-Pd(II) Verses Pd(II) - Pd(IV) 176 but later ruled

out the Pd(IV) intermediate for phosphapalladacycles177.R

Na(

NaX HO Ac

HIX - P d - L

I

R

L Vii i

Figure 1.32: The generally accepted Heck reaction pathway.


On examining the mechanism for the Heck reaction it can be seen that it is composed

of several fundamental steps, preactivation of the catalyst, oxidative addition,

migratory insertion, (3-hydride elimination and reductive elimination. In order to form

(I), Pd(II) complexes must first be reduced to Pd(0) by employment of suitable

reducing agents such as hydrazine or sodium formate, failure of this leads to ani ininductive phase . The actual catalytic species (I), assumed to be a coordinatively

unsaturated 14e- Pd(0) species168,169,178, is generated by the coordination of acetate

anions. This was demonstrated by Armatore and Jutand for Pd/PAr3 systems who

showed that acetate anions, present from the NaOAc employed as base, underI I n o

catalytic conditions coordinate to the Pd centre to give an anionic Pd(0) species ’ .

This also provides a good explanation for the hydride removal from the Palladium.

Pd(0) complexes may also be used as the catalyst eliminating the reduction step,

though ligand dissociation from the pre-catalyst may be required to form the

coordinatively unsaturated active catalyst. Oxidative addition of an aryl halide then

takes place with (I) to give (II) as a concerted process where the C-X bond breaks and

corresponds with the formation of M-C and M-X bonds. The reactivity of the aryl

halide decreases drastically in the order of I > Br > Cl179. The strength of these bonds

in Caryi-X compounds (kJ mol'1) are PhCl (402), PhBr (339) and Phi (272)180. The cis

isomer is also formed in this process, though only the thermodynamically more stable

trans-isomer is shown in the catalytic cycle. Dissociation of a ligand or halide then

provides a vacant coordination site for which the olefin can attach to give (III),

followed by olefin insertion into the Pd-aryl bond to give (IV). A (3-hydride

elimination process then occurs to give the coupled product thereby leaving the Pd

species to undergo reductive elimination in the presence of base to regenerate the

initial Pd(0) species (I).

1.7.3 NHCs As Efficient Ligands For The Heck Reaction

Herrmann first demonstrated the advantages of NHC’s over their traditional

phosphine counterparts for the Heck reaction in 1995. Generally in traditional systems

using triarylphosphines an excess of the phosphine ligand is needed to control the

equilibrium in the activation and propagation steps of the catalytic cycle130 due to


phosphane degradation by P-C bond cleavage181. This excess contributes to higher

running costs in technical plants. Other disadvantages of phosphines are their

sensitivity towards air and moisture. The inherent stability of the M-NHC bond gives

rise to NHC catalysts that are active for long periods at elevated temperatures, often

without the need for specially purified solvents or inert atmospheres. Herrmann’s

Pd(II) complexes (1.64, 1.65) were shown to be extraordinarily stable to heat, oxygen

and water130, and under catalytic conditions 1.64 was shown to give high TON’s as a

result of its pronounced thermal stability in solution. Herrmann concluded that the

catalytic advantages of complex 1.64, was that it could form stable active species for

long periods of time and at high temperatures which are essential properties for the1 inactivation of chloroarenes . In the same paper Herrmann prepared a [Pd(0 )(carbene)2]

complex by the addition of two equivalents of free l,3-dimethyldihydroimidazole-2-

ylidene with [Pd(dba)2] and demonstrated that if the initial Pd complex is a Pd(0) one,

then an induction period does not occur. This lack of induction period coupled with

the high activity due to the carbene ligand resulted in high TON’s (in comparison to

non NHC Pd(0) complexes). These findings led to a number of groups synthesising

and testing mono PdNHC complexes4 9 ,117,182-184 in the Heck (and related) reaction(s).

Further work carried out in this area with chelating, functionalised carbenes has

extended the range of substrates used as well as demonstrated how ligand design can'y 'y c / : “j a i a q i o c i q a

affect overall activities ’ ’ ’ ’ " .A number of pyridine functionalised carbene

complexes of Pd have been shown to be active catalysts for the Heck reaction (with

some examples given in previous section of this chapter) and are relevant to the

complexes synthesised within this work56,74,75,1 182 ,189

Pd

1.64

Pd

1.65

Figure 1.33: Herrmann’s first NHC Pd complexes.


1.8 Aims Of This Thesis

The main aim of this work is the synthesis of novel chelating carbene complexes of

the catalytically interesting Pd metal. The investigation of the catalytic abilities of the

complexes is also of priority. The ligands formed herein were designed with the aim

to contain a combination of features that could lend an advantage in catalytic systems.

These were chosen to be chelation, rigidity, hemilability and chirality.

Chapter 2 describes the synthesis and characterisation of a number of novel

imidazolium salts encompassing bis-imidazolium salts and functionalised

imidazolium salts.

Chapter 3 gives the synthesis and characterisation of Ag(I)NHC complexes of the

imidazolium salts described in chapter 2 , which were synthesised for the purpose of

being intermediate transmetallation agents for the synthesis of the desired Pd(II)

complexes of chapter 4.

The successful transfer from Ag(I) to Pd(II) of a number of the Ag complexes of

chapter 3 are discussed in chapter 4. Both the di-chloro and the catalytically important

hydrocarbyl complexes are synthesised for a number of ligands, and are characterised

and supported by crystallographic evidence.

Chapter 5 presents catalytic results for a selection of the Pd complexes of chapter 4 in

the Heck reaction of 4-bromoacetophenone and n-butyl acrylate, as well as 4-

chlorobenzaldehyde with n-butyl acrylate. The results are compared to complexes

with similar structural features previously synthesised by McGuinness56, and are

shown to be comparable with respect to the coupling of 4-bromoacetophenone and

inferior with respect to 4-chlorobenzaldehyde.


1.9 References1. D. Bourissou, O. Guerret., F. Gabbai, G. Bertrand, Chem. Rev., 2000.100. 39-

91.

2. G. Schuster, Advanced Physical Organic Chemistry, 1986,22, 311.

3. R. Hoffmann.,./ Am. Chem. Soc., 1968, 90, 1475.

4. E. O. Fischer, A. Maasbol., Angew. Chem. Int. Ed. Engl., 1964,3, 580.

5. H. Wanzlick, E. Schikora., Chem. Ber., 1961, 94,2389-2393.

6 . H. Wanzlick, Angew. Chem. Int. Ed. Engl., 1962,1, 75-80.

7. H. Wanzlick, B. Lachmann., E. Schikora, Chem. Ber., 1965,98, 3170-3177.

8 . V. Bohm, W. Herrmann, Angew. Chem. Int. Ed. Engl., 2000,39, 4036-4038.

9. H. E. Winberg, J. E. Carnahan, D. D. Coffman, M. Brown, J. Am. Chem. Soc.,

1965, 87, 2055.

10. D. M. Lemal, R. A. Lovald, K. I. Kawano, J. Am. Chem., 1964, 8 6 , 2518.

11. F. E. Hahn, L. Wittenbecher, D. L. Van, R. Frohlich., Angew. Chem. Int. Ed.,

2000,39, 3, 541.

12. W. Herrmann, V. Bohm, Angew. Chem. Int. Ed., 2000,39, 22,4036.

13. D. Lemal, Y. Liu, P. Lindner, J. Am. Chem. Soc., 1999,121, 10626.

14. A. Arduengo, R. Harlow, M. Kline, J. Am. Chem. Soc., 1991,113, 361.

15 A. J. Arduengo, H. Dias, R. L. Harlow, M. Kline, J. Am. Chem. Soc., 1992,

114, 5530.

16. D. Ender, K. Breuer, G. Raabe, J. Runsink, J. Teles, J. Melder, K. Ebel, S.

Brode, Angew. Chem. Int. Ed., 1995. 34, 9, 1021-1023.

17. A. Arduengo, J. Goerlich, W. Marshall, Liebigs Ann., 1997, 365-374.

18. R. Alder, M. Blake, C. Bortolotti, S. Bufali, C. Butts, E. Linehan, J. Oliva, A.

Orpen, M. Quayle, Chem. Comm., 1999,3, 241.

19. D. Martin, A. Baceiredo, H. Gomitzka, W. Schoeller, G. Bertrand, Angew.

Chem. Int. Ed., 2005, 44, 11, 1700-1703.

20. V. Lavallo, Y. Canac, C. Prasang, B. Donnadieu, G. Bertrand, Angew. Chem.

Int. Ed., 2005,44, 5705-5709.

21. R. Alder, P. Allen., M. Murray, A. Orpen, Angew. Chem. Int. Ed., 1996, 35, 10,

1121-1123.

22. R. Alder, C. Butts, A. Orpen, J. Am. Chem. Soc., 1998,120, 11526-11527.


23. V. Lavallo, J. Mafhouz, Y. Canac, B. Donnadieu, W. Schoeller, G. Bertrand., J.

Am. Chem. Soc., 2004,126, 8670.

24. N. Clement, Ph.D. Thesis, 2005.

25. M. Palucki, S. Buchwald, J. Am. Chem. Soc., 1997, 119, 11108

26. B. Hamann, J. Hartwig, J. Am. Chem. Soc., 1997 119, 12382

27. T. Satoh, Y. Kawamura, M. Miura, M. Nomura, Angew. Chem. Int. Ed. Eng.,

1997, 36, 1740.

28. V. Lavello, Y. Canac, B. Donnadieu, W. Schoeller, G. Bertrand, Angew. Chem.

Int. Ed, 2006,45, 3488-3491.

29. D. Dixon, A. Arduengo, K. Dobbs, D. Khasnis, Tetrahedron Lett., 1995, 36,

645

30. K. Ofele, W. Herrmann., D. Mihalios, M. Elison, E. Herdtweck, W. Scherer, J.

Mink, J. Organomet. Chem., 1993,459, 177-184.

31. J. Huang, H. Schanz, E. Stevens, S. Nolan, Organometallics, 1999,18, 2370-

2375.

32. J. Schawarz, V. Bohm, M. Gardiner, M. Groshe, W. Herrmann, W. Hieringer,

G. Raudaschl-Sieber, Chem. Eur. J., 2000, 6 , 1773.

33. W. Herrmann, T. Weskamp, V. Bohm, Adv. Organomet. Chem., 2001, 48, 1-

69.

34. J. Green, R. Scurr, P. L. Arnold, F. G. Cloke, Chem. Comm., 1997,20, 1963-

1964.

35 P. Arnold, S. Mungur, A. Blake, C. Wilson, Angew Chem. Int. Ed. 2003, 42,

5981-5984

36. A. Arduengo, M. Tam, J. Calabrese, F. Davidson, W. Marshall, Chem. Lett.,

1999, 1021-1022.

37. W. Herrmann, O. Runte, G. Artus, J. Organomet. Chem., 1995, 501, C1-C4.

38. A. Arduengo, M. Tam, S. McLain, J. Calabrese, F. Davidson, W. Marshall, J.

Am. Chem. Soc, 1994,116, 7927.

39. W. J. Oldham, S. M. Oldham, B. L. Scott, K. D. Abney, W. H. Smith, D. A.

Costa, Chem. Comm., 2001, 15, 1348.

40. P. Arnold, Heteroat. Chem., 2002,6 , 534.

41. J. Chen, I. Lin, Organometallics, 2000,19, 5113-5121.

42. C. Crudden, D. Allen., Coord. Chem. Rev., 2004, 248, 2247-2273.


43. D. McGuinness, N. Saendig, B. Yates, K. Cavell, J. Am. Chem. Soc, 2001,123,

4029-4040.

44. S. Caddick, G. Cloke, P. Hitchcock, J. Leonard, A. Lewis, D. Mckerrecher, L.

Titcomb, Organometallics, 2002,21, 4318-4319.

45. D. McGuinness, W. Mueller, P. Wasserscheid, K. Cavell, B. Skelton, A.

White, U. Englert, Organometallics, 2002,21, 175-181.

46. W. Marshall, V. Grushin, Organometallics, 2003,22, 1591.

47. K. Cavell, D. McGuinness, Coord. Chem. Rev., 2004, 248, 671-681.

48. D. McGuinness, V. Gibson, J. Stead, Organometallics, 2004, 23, 6288-6292.

49. D. McGuinness, K. Cavell, M. Green, B. Skelton, A. H. White, J. Organomet.

Chem., 1998, 565, 165.

50. H. Dias, W. Jin, Tetrahedron Let., 1994,35, 1365

51. M. Anderson, A. Casalnuovo, B. Johnson, B. Mattson, A. Mueting, L.

Pignolet, Inorg. Chem., 1988, 27, 1649.

52. Z. Zhang, H. Chen, Coord. Chem. Rev., 1996,147, 1.

53. A. Aranyos, D. Old, A. Kiyomori, J. Wolfe, J. Sadighhi, S, Buchwald, J. Am.

Chem. Soc., 1999,121, 4369.

54. K. Reddy, C.Chen., Y. Liu, S. Peng, J. Chen, S. Liu, Organometallics, 1999,

18, 2574.

55. C. Amatore, A. Fuxa, A. Jutand, Chem. Eur. J., 2000, 8, 1474.

56. D. McGuinness, K.Cavell, Organometallics, 2000,19, 741.

57. J. Chen, I. Lee, J. Chem. Soc., Dalton Trans., 2000, 839.

58. X. Wang, S. Liu, G. Jin, Organometallics, 2004, 23, 6002-6007

59. S. Mungur, S. Liddle, C. Wilson, M. Sarsfield, P. Arnold, Chem Comm, 2004,

2738-2739

60. S. Liddle, P. Arnold, Organometallics, 24, 2597-2605.

61. R. Douthwaite, J. Houghton, B. Kariuki, Chem Comm, 2004, 698-699

62. M. Froseth, A. Dhindsa, H. Roise, M. Tilset, Dalton Trans, 2003, 4516-4524.

63. K. Coleman, H. Chamberlayne, S. Turberville, M. Green, A. Cowley, Dalton

Trans, 2003, 2917-2922

64. P. Arnold, M. Rodden, K. Davis, A. Scarisbrick, A. Blake, C. Wilson, Chem.

Comm., 2004, 1612-1613

65. P. Arnold, A. Scarisbrick, A. Blake, C. Wilson, Chem. Comm., 2001, 2340-

2341.


6 6 . P. Arnold, A. Scarisbrick, Organometallics, 2004, 23, 2519-2521.

67. J. Veldhuizen, S. Garber, J. Kingsbury, A. Hoveyda, J. Am. Chem. Soc., 2002,

124, 4954-4955

6 8 . A. Waltmann, R. Grubbs, Organometallics, 2004, 23, 3105-3107

69. C. Yang, H. Lee, S. Nolan, Org. Lett., 2001,3, 10, 1511

70. N. Tsoureas, A. Danopoulos, A. Tulloch, M. Light, Organometallics, 2003, 22,

4750.

71. A. Tulloch, S. Winston, A. Danopoulos, G. Eastham, M. Hursthouse, J. Chem.

Soc. Dalton Trans., 2003, 699-708

72. R. Douthwaite, M. Green, P. Silcock, P. Gomes, J. Chem. Soc., Dalton Trans.,

2002,1386-1390.

73. W. Herrmann, C. Kocher, Angew. Chem. Int. Ed., 1997,36, 2162-2187.

74. E. Peris, J. Loch, J. Mata, R. Crabtree, Chem. Comm., 2001, 201.

75. J. Loch, M. Albrecht, E. Peris, J. Mata, J. Faller, R. Crabtree, J. Am. Chem.

Soc.,2002,21(4) 700-706.

76. M. Poyatos, J. Mata, E. Falomio, R. Crabtree, E. Peris, Organometallics, 2003,

22, 1100-1114

77. D. J. Nielsen, K. Cavell, B. W. Skelton, A. H. White, Inorg. Chim. Acta, 2002,

327, 116.

78. A. Danopoulos, A. Tulloch, S. Winston, G. Eastham, M. Hursthouse, Dalton

Trans., 2003, 1009-1015.

79. A. Danopoulos, N. Tsoureas, J. Wright, M. Light, J. Am. Chem. Soc., 2004,

23(2) 166-168

80. A. Danopoulos, J. Wright, W. Motherwell, Chem. Commun., 2005, 784-786.

81. A. Danopoulos, J. Wright, W. Motherwell, S. Ellwood, Organometallics, 2004,

23,4807-4810

82. D. McGuinness, V. Gibson, D. Wass, J. Steel, J. Am. Chem. Soc., 2003,125,

12716-12717.

83. H. Lee, J. Zeng, C. Hu, M. Lee, Inorg., Chem., 2004, 43, 6822-6829.

84. A. Melaiye, R. Simons, A. Milsted, F. Pingitore, C. Wesdemiotis, C. Tessier,

W. Youngs, J. Med. Chem., 2004, 47, 973-977.

85. M. Perry, K. Burgess, Tetrahedron: Asymmetry, 2003,14, 951.

8 6 . V. Cesar, S. Bellemin-Laponnaz, L. Gade, Chem. Soc. Rev., 2004, 33, 619-639

87. J. Whitesell, Chem. Rev., 1989, 89, 1581-1590.


8 8 . W. Herrmann, L. Goossen, C. Kocher, G. Artus, Angew Chem. Int. Ed. Engl.,

1996, 35, 2805.

89. D. Enders, H. Gielen, K. Bever, Tetrahedron: Asymmetry, 1997, 8 , 3571.

90. S. Lee, J. Hartwig, J. Org. Chem., 2001, 6 6 , 3402.

91. M. Powell, D. Hou, M. Perry, X. Cui, K. Burgess, J. Am. Chem. Soc., 2001,

123, 8878-8879.

92. R. Broene, S. Buchwald, J. Am. Chem. Soc., 1993, 115, 12569-12570.

93. J. Blankenstein, A. Pfaltz, Angew Chem. Int Ed., 2001,40,4445-4447.

94. F. Menges, A. Pfaltz, Adv. Syn. Catal., 2002, 344,40-44.

95. J. Pytkowicz, S. Roland, P. Mangeney, Tetrahedron: Asymetry, 2001,12,

2083.

96. (a) F. Guilten, C. Winn, A. Alexakis, Tetrahedron: Asymetry, 2001,12, 2083.

(b) A. Alexakis, C. Winn, F. Guillen, J. Pytkowickz, S. Roland, P. Mangeney,

Adv. Synth. Catal., 2003, 3, 345.

97. T. Seiders, D. Ward, R. Grubbs, Org. Lett., 2001,3, 3225.

98. J. Wolfe, S. Nagaw, J. Marcoux, S. Buchwald, Acc. Chem. Res., 1998, 31, 805.

99. E. Bappert, G. Helmchen, Synlett., 2004, 1789.

100. D. Clyne, J. Jin, E. Genest, J. Gallucci, T. Rajanbabu, Org. Lett., 2000,2,

1125.

101. W. Duan, M. Shi, G. Rong, Chem. Comm., 2003, 2976.

102. C. Bolm, M. Kesselgruber, G. Raabe, Organometallics, 2002, 21, 707.

103. M. Perry, X. Cui, K. Burgess, Tetrahedron: Asymmetry, 2002,13, 1969.

104. L. Bonnet, R. Douthwaite, B. Kariuki, Organometallics, 2003,22, 4187.

105. W. Herrmann, L. Goossen, M. Spiegler, Organometallics, 1998,17, 2162.

106. (a) M. Powell, D. Hou, M. Perry, X. Cui, K. Burgess, J. Am. Chem. Soc., 2001,

123, 8878. (b) M. Perry, X. Cui, M. Powell, D. Hou, H. Reibenspies, K.

Burgess, J. Am. Chem. Soc., 2003, 125, 113.

107. A. Arduengo, Acc. Chem. Res., 1999,32, 913.

108. T. Weskamp, V. Bohm, W. Herrmann, J. Organomet. Chem., 2000, 600, 12-22.

109. A. Tulloch, A. Danopoulos, R. Tooze, S. Cafferkey, S. Kleinhenz, M.

Hursthouse, Chem. Comm., 2000,14, 1247-1248.

110. J. Loch, M. Albrecht, E. Peris, J. Mata, J. Faller, R. Crabtree, Organometallics,

2002, 21, 700-706.


111. A. Danopoulos, S. Winston, T. Gelbrich, M. Hursthouse, R. Tooze, Chem.

Comm., 2002, 482-483.

112. A. Danopoulos, S. Winston, W. Motherwell, Chem. Comm., 2002,1376-1377.

113. W. Herrmann, M. Elison, J. Fischer, C. Kocher, G. Artus, Chem. Eur. J. 1996,

2, 772

114. W. Herrmann, C. Kocher, L. Gooben, G. Artus, Chem. Eur. J., 1996, 2,1627.

115. W. Fehlhammer, T. Bliss, U. Kembach, L. Brudgam, J. Organomet. Chem.,

1995, 490, 149.

116. N. Kuhn, T. Kratz, Synthesis, 1993, 561.

117. D. S. McGuinness, K. Cavell, B. Skelton, A. H. White, Organometallics, 1999,

18, 1596.

118. A. Arduengo, S. Gamper, J. Calabrese, F. Davidson, J. Am. Chem. Soc., 1994,

116, 4391-4394.

119. V. Bohm, W. Herrmann, C. Gstottmayr, T. Weskamp, J. Organomet. Chem.,

2000, 595, 186-190.

120. K. Selvarkumar, A. Zapf, A. Spannenberg, M. Beller, Chem. Eur. J., 2002, 8 ,

3901-3906.

121. M. Green, K. Cavell, B. Skelton, A. H. White, J. Organomet. Chem., 1998, 554,

175-179.

122. M. Viciu, R. Germaneau, O. Navarro-Femandez, E. Stevens, S. Nolan,

Organometallics, 2002, 21, 5470-5472.

123. M. Viciu, R. Kissling, E. Stevens, S. Nolan, Organic Letters, 2002, 4, 2229-

2231.

124. H. Wanzlick, H. Kleiner, Angew. Chem., 1961, 79,493.

125. J. Chamizo, J. Morgado, S. Bernes, Transition Met. Chem., 2000,25, 161-165.

126. M. Scholl, S. Ding, C. Lee, R. H. Grubbs, Organic Letters, 1999,1, 953-956.

127. J. Teles, J. Melder, K. Ebel, R. Schneider, E. Gehrer, W. Harder, S. Brode, D.

Enders, K. Breuer, G. Raabe, Helv. Chim. Acta, 1996, 79, 61-83.

128. K. Ofele, J. Organomet. Chem., 1968,12,42.

129. H. Wanzlick, H. Schonherr, Angew. Chem. Int. Ed., 1968, 7, 141.

130. W. Herrmann, M. Elison, J. Fischer, C. Kocher, G. Artus, Angew. Chem. Int.

Ed. Eng., 1995,34,2371.

131. W. Herrmann, A. Fischer, K. Ofele, G. Artus, J. Organomet. Chem., 1997, 530,

259.


132. W. Herrmann, C. Reisinger, M. Spiegler, J. Organomet. Chem., 1998, 557, 93.

133. M. Gardner, W. Herrmann, C. Reisinger, J. Schwarz, M. Spiegler, J.

Organomet. Chem., 1999, 572,239.

134. W. Herrmann, J. Schwarz, M. Gardner, M. Spiegler, J. Organomet. Chem.,

1999, 575, 80

135. W. Herrmann, J. Schwarz, M. Gardner, Organometallics, 1999,18,4082.

136. H. Lebel, M. Janes, A. Charette, S. Nolan, J. Am. Chem. Soc., 2004,126, 5046.

137. D. McGuinness, K. Cavell, B. Yates, B. Skelton, A. White, J. Am. Chem. Soc.,

2001,123, 8317.

138. A. Furstner, G. Seidel, D. Kremzow, C. W. Lehmann, Organometallics, 2003,

22, 907.

139. M. S. Viciu, G.A.Grasa., S. P. Nolan, Organometallics, 2001,20, 3607.

140. S. Grundemann, M. Albrecht, A. Koacevic, J. W. Faller, R. H. Crabtree, J.

Chem. Soc., Dalton Trans., 2002,10, 2163.

141. N. Clement, K. Cavell, C. Jones, C. Elsevier, Angew. Chem. Int. Ed., 2004, 43,

1277-1279.

142. S. Liu, T. Hsieh, S. Peng, Organometallics, 1998,17, 993.

143. R. Ku, J. Huang, J. Cho, F. Kiang, K. Reddy, Y. chen, K. Lee, J. Lee, G. Lee,

S. Peng, S. Liu, Organometallics, 1999,18, 2145.

144. P. Arnold, F. Cloke, T. Geldbach, P. Hitchcock, Organometallics, 1999, 3228-

3233.

145. Heaton, A., An introduction to industrial chemistry. 3 ed. 1996: Blackie A & P.

146. C. Elschenbrioch, A. Salzer, Organometallics: 2nd Revised Ed. VCH:

Wurzburg, 1992,411

147. B. Comils, W.H., Applied homogeneous catalysis with organometallic

compounds. 1996: VCH:Weinheim.

148. D. Nielsen, PhD Thesis, 2002.

149. A. C. Hillier, G. A. Grasa, M. S. Viciu, H. M. Lee, C. Yang, S. P. Nolan, J.

Organomet. Chem., 2002, 653, 69.

150. K. Tamao, Y. Kiso, K. Sumitani, M. Kumada, J. Am. Chem. Soc., 1972, 94,

9268.

151. A. Sekia, N. Ishikawa, J. Organomet. Chem., 1976,118, 349.

152. E. Erdik, Tetrahedron, 1992, 48, 95, 77.

153. J. K. Stille, Angew. Chem. Int. Ed. Engl., 1986, 25, 508.


154. A. Zapf, M. Beller, Chem. Commun., 2005,431.

155. N. Miyaura, A. Suzuki, Chem. Rev., 1995, 95, 2457.

156. A. Zapf, R. Jackstell, F. Rataboul, T. Riermeier, A. Monsees, C. Fuhrmann, N.

Shaikh, U. Dingerdissen, M. Beller, Chem. Commun., 2004,1, 38.

157. M. Huser, M. Youinou, J. Osborn, Angew. Chem. Int. Ed. Eng., 1989,28,

1386.

158. W. Herrmann, C. Brobmer, T. Priermeier, K. Ofele, J. Organomet. Chem.,

1994, 481, 97.

159. J. Wolfe, S. Wagaw, J.-F. Marcoux, S. L. Buchwald, Acc. Chem. Res., 1998,

31, 805.

160. J. F. Hartwig, Acc. Chem. Res., 1998,31, 852.

161. J. F. Hartwig, Angew. Chem. Int. Ed. Engl., 1998,37, 2046.

162. J. F. Hartwig, Synlett, 1997, 329.

163. B. H. Yang, S. L. Buchwald, J. Organomet. Chem., 1999, 576, 125.

164. I. Beletskaya, A. Cheprakov, Chem. Rev., 2000,100, 3009-3066.

165. W. A. Herrmann, Angew. Chem. Int. Ed., 2002, 41, 1290.

166. E. Peris, R. H. Crabtree, Coordination Chemistry Reviews, 2004,248, 2239.

167. A. Littke, G. Fu, J. Am. Chem. Soc., 2001,123,6989-7000.

168. A. Schnyder, A. Indolese, M. Studer, H. Blaser, Angew. Chem. Int. Ed., 2002,

41, 3668-3671.

169. G. Li, Angew. Chem. Int. Ed., 2001, 40, 1513-1515.

170. C. Amatore, A. Jutand, Acc. Chem. Res., 2000,33, 314-321.

171. P. L. Arnold, F. G. N. Cloke, T. Geldbach, P. B. Hitchcock, Organometallics,

1999,18, 3228.

172. K. Albert, P. Gisdakis, N. Roesch, Organometallics, 1998,17, 1608.

173. B. L. Shaw, S. P. Perera, E. A. Staley, J. Chem. Soc. Chem Commun., 1998,

13, 1361.

174. W. Herrmann, C. Brossmer, K. Ofele, C. Reisinger, T. Priermeier, M. Beller,

H. Fischer, Angew. Chem. Int. Ed. Eng., 1995,34, 1844-1848.

175. D. Morales, R. Redon, C. Yung, C. Jensen, Chem. Comm., 2000, 1619-1620.

176. W. A. Herrmann, V. P. W. Bohm, C.-P. Reisinger, J. Organomet. Chem., 1999,

576, 23-41.

177. V. Bohm, W. Herrmann, Chem. Eur. J., 2001, 7(19), 4191-4197.

178. C. Amatore, A. Jutand, J. Organomet. Chem., 1999, 576, 254.


179. N. Gurbuz, I. Ozdemir, S. Demir, B. Cetinkaya, J. Molec. Cat. A, Chem., 2004,

209, 23-28

180. M. B. Andrus, C. Song, Org. Lett., 2001,3, 3761.

181. P. Garrou, Chem. Rev., 1985, 85, 171-185.

182. A. Magill, D. McGuinness., K. Cavell, G. Britovsek, V. Gibson, A. White, D.

Williams, A. H. White, B. Skelton, J. Organomet. Chem., 2001, 546, 617-618.

183. W. Herrmann, V. Bohm, C. Gstottmayr, M. Grosche, C. Reisinger, T.

Weskamp, J. Organomet. Chem., 2001, 617-618, 616-628.

184. M. Baker, B. Skelton, A. White, C. Williams, J. Chem. Soc., Dalton trans.,

2001, 111- 120.

185. S. Grundemann, M. Albrecht, J. Loch, J. Faller, R. Crabtree, Organometallics,

2001,20,5485.

186. J Loch, R. Crabtree, Pure Appl. Chem., 2001, 73, 119-128.

187. R. Batey, T. Quach, M. Shen, A. Thadani, D. Smil, S. Li, D. Mackay, Pure

Appl. Chem., 2002, 74, 43-55.

188. R. Batey, M. Shen, A. Lough, Organic Letters, 2002, 4 ,1411-1414.

189. A. Tulloch, A. Danopoulos, M. Hursthouse, G. Tizzard, S. Coles, R. Hay-

Motherwell, W. Motherwell, Chem. Comm., 2001, 1270.

190. V. Cesar, S. Bellemin-Laponnaz, L. Gade, Organometallics, 2002,21, 5204-

5208.

191. K. Selvarkumar, A. Zapf, M. Beller, Organic Letters, 2002, 4, 3031-3033.

192. T. Jeffery, J. Chem. Soc., Chem. Comm., 1984, 1287.

193. (a) M. Lappert, J. Organomet. Chem. 1975, 100, 139-159. (b) M. Lappert, P.

Pye, J. Chem. Soc. Dalton Trans. 1977, 21, 2172-2180. (c) M. Lappert, J.

Organomet. Chem. 1988,358, 185-213.

Chapter Two: Chelating Imidazolium Salts

Chapter Two

Ligand Design: Chelating Imidazolium Salts And

Their Synthesis

2.1 Introduction.

Imidazolium salts are not the only route to forming carbenes , though they are used

throughout this work to form the NHC’s and associated metal complexes. Herrmann’s

description of an imidazolium salt is a 5 membered heterocycle with nitrogens at the 1

and 3 positions and a substituent (H, R, Ar or X) at each position of the ring. The ring

members are sp hybridized and the ring bears a single positive charge that is

delocalised around the ring. 5 A typical imidazolium salt is shown in (I) (Figure 2.1).

R3 R3 R3

(II) ("I) (IV)

Figure 2.1: N-Heterocyclic free carbenes and their corresponding salts

The numbering shown in (I) will be used throughout this work and unless otherwise

stated, R.2 = R4 = R5 = H. Free carbenes (II) and (IV) are derived from (I) and (III)

respectively and are named according to the degree of saturation on the C 4 - C 5


backbone, i.e. (II) is an imidazolin-2-ylidene and (IV) an imidazolidin-2-ylidene. The

work contained within this thesis will consist of types (I) and (II).

Chapter one noted that a chelate forming ligand can be especially beneficial in terms

of offering stability to hydrocarbyl-M-carbene complexes with respect to reductive

elimination6. The general theme of this chapter is thus based upon the design and

synthesis of imidazolium salts as chelating carbene precursors with the majority of the

imidazolium salts synthesised being functionalized mono-imidazolium salts with one

example of a bis imidazolium salt. Two of the synthesised imidazolium salts also

contain chiral centres.

A new chelating, rigid mixed donor ligand was desired that could be comparative to

previous ligands and complexes synthesised by Cavell (2.1 and 2.2, Scheme 2.1)7 and

others that were effectively employed in the Heck reaction. Chelating ligands, and in

particular those having both strong and weak donors (hemilabile ligands), are

particularly interesting in terms of catalysis. The weak hemilabile part of the ligand is

capable of reversible dissociation from the metal centre, thereby creating vacant

coordination sites during catalytic cycles and stabilising the metal centre by re

coordination when it is catalytically non active, or coordinatively unsaturated.

Nal

I-

Ag201/2 ,Ag(|)(carbene)2'

PdMeCI(COD)

2.1 2.2

Figure 2.1: Reaction scheme showing the synthesis of Cavell’s mixed donor

ligand precursor and its corresponding Pd complex.


The majority of functionalised chelating carbenes have utilized (CH2)n chains as

linkers to bridge the NHC and secondary donors together which are highly convenient

from a synthetic viewpoint. A selection of these are given in Figure 2.2. However it

has become apparent that there are drawbacks in that the CH2 protons can be too

acidic and may be abstracted by the strong bases traditionally used in the

deprotonation step for free carbene formation . Using an aromatic ring system to

bridge the donors will overcome this, and so an imidazolium salt with a quinoline

structure as one of the N-substituents (Figure 2.3) was thought to be an ideal

candidate. The imidazolium salt also has the added feature that it is a highly

delocalized system extending over three rings. In order to assess the effect of sterics in

the ligands associated complexes, a secondary objective in forming these salts is the

manipulation of steric bulk of the R- group (Figure 2.3) as well as the addition of a

group on the 2 position of the quinoline moiety (Figure 2.3) which may lead to

increased hemilability.

(d)12 (f)14

Figure 2.2: A selection of functionalised NHC precursors with their donors bridged by

CH2(n> linkers9"14.


2.2 Results and discussion

2.2.1 Preparation of functionalized imidazolium salts.

2.2.1.1 Quinoline based imidazolium salts.

5 4

6

7

11

12

R

Figure 2.3: Target quinoline based salt including the numbering scheme for the

quinoline moiety.

The commercial availability of 8-aminoquinoline and 2-methyl-8-nitroquinoline

removes the need for a synthetic method for which to construct a quinoline based

compound suitable for the salt synthetic routes outlined in Chapter one, thereby

reducing the total number of synthetic steps for imidazolium salt synthesis.

2-Methyl-8-aminoquinoline (2.3, Scheme 2.2) was synthesised by vigorously stirring

a solution of 2-methyl-8-nitroquinoline (less than 0.5g in ca. 20ml THF) with catalytic

amounts of 10% Pd on carbon under an atmosphere of H2 overnight. The reaction

mixture was then filtered through celite and the solvent removed under reduced

pressure. !H NMR indicated that the reaction had gone to completion, observed by a

broad peak at around 4.5-5ppm corresponding to the N//2 . 2-methyl-8-aminoquinoline

was used for the imidazole synthesis without further purification. As this reaction was

successful, this method was preferred over other reduction methods, such as

zinc/hydrazinium monoformate, despite the long periods of time required to convert

significant amounts of the starting material.


Ho, 10% Pd on C

2.3

Scheme 2.2: Reduction of 2-methyl-8-nitroquinoline.

The synthetic strategy for the imidazole ring synthesis of 2.5 was adopted from

Zhang’s modified procedure for the synthesis of 1-arylimidazoles15 such as 2.4. The

imidazolium salts were formed by standard N-alkylation using methyl iodide or

benzyl bromide (Scheme 2.3).

Mel or PhCH,BrFunctionalised Imidazolium Salts

Scheme 2.3: Stepwise imidazolium salt synthesis.

The formation of the 1-aromatic substituted imidazoles (2.4 and 2.5) proceeded

according to the literature preparative method15 (Scheme 2.3), although these

reactions both suffered from low yields (-30%) which is a recognised problem in the

literature with most 1-aromatic substituted imidazoles16,17. The generation of large

amounts of an unknown material as well as inorganic salts during neutralization

proved to be problematic and may have contributed to the low yields. Vigorous

agitation and copious volumes of Et2 0 during the extraction of the amine marginally

improved product yield. Smaller scale reactions were more efficient in terms of yields,

although overall yields were highly variable.

The second R group was limited to methyl and benzyl due to the readily availability

and reactivity of the respective alkyl halides as N-alkylation of the imidazole ring

follows typical Sn2 reactivity, i.e. quartemization is difficult to achieve with any


nucleophile less reactive than a secondary alkyl bromide (though not impossible). The

functionalized imidazoles were reacted with either methyl iodide or benzyl bromide in

THF overnight to give the required imidazolium salts as light brown solids which

show good stability towards air and moisture. The following imidazolium salts havei nbeen prepared (Figure 2.4), and were characterised by H, C NMR and Mass

Spectrometry (MS). The characteristic peak in the ]H NMR is the imidazolium proton

(Position 14, Figure 2.3) as a singlet at 10.15-10.75ppm.

Br-

2.9

Figure 2.4: The range of quinoline functionalised imidazolium salts

synthesised within this work.

The inclusion of a methyl group at the 2 position of the quinoline moiety (salts 2.8

and 2.9) offers additional steric bulk which satisfies one of the objectives of this

section and may have an effect of increasing the lability of the donor group by

sterically crowding the coordination sphere. This would be advantageous in some

catalytic cycles where hemilability contributes both to catalyst stabilization and the

creation of vacant coordination sites.

Crystals of 1-methyl-3-(2-methyl-quinoline)-imidazolium iodide (2.8) suitable for X-

ray crystallography were grown by layering hexane onto a DCM solution of the salt.

The structure and labelling scheme of the cation of 2.8 is shown in Figure 2.5 along

with a packing diagram for the salt (Figure 2.6). Selected bond lengths and angles are

given in Table 2.1.


C9C7

C8C 10C6

C5C 12C11C 4

N3C 13

N2

C 3

C2N1

C 14

Figure 2.5: ORTEP representations of 1-methyl-3-(2-methyl-quinoline)-imidazolium iodide (2.8) (50% probability of thermal ellipsoids). Hydrogen and iodide atoms are

omitted for clarity.


Figure 2.6: ORTEP packing representation of 1-methyl-3-(2-methyl-quinoline)-

imidazolium iodide (2.8) (Hydrogen atoms omitted for clarity).

Cl N1 1.326(4) N1 C14 1.460(5) N1 Cl N2 108.8(3)

Cl N2 1.336(4) N2 C4 1.439(4) Cl N1 C14 126.0(3)

C2 C3 1.340(6) C4 C12 1.424(4) Cl N2 C4 127.5(3)

C2 N1 1.376(4) C12 N3 1.360(4) N3 C12 C4 119.7(3)

C3 N2 1.397(4)

Table 2 .1 : Selected bond lengths (A) and angles (°) of 2 . 8

The planes of the imidazolium and quinoline rings make an angle of approximately

33.1°. The imidazole ring is typical of others with an N1-C1-N2 bond angle of

108.8(3)°, and bond lengths of 1.326(4)A and 1.336(4)A respectively18'19. The

packing diagram shows n-n interactions between layers of stacked quinoline moieties.

It could be envisaged that if the ligand were to form a chelated complex, a biaxial

chirality could be present by this deviation from co-planarity. In an attempt to exploit

this, it was thought that by increasing the steric bulk on the C3 and C5 positions


(Figure 2.5) it may be possible to increase the torsional angle between the two planes

and increase the energy barrier of interconversion. An aromatic carbene backbone was

thought to be an ideal candidate, and so a suitable synthetic method was sought. A

method was found in the literature and was successfully applied with 8 -

aminoquinoline to give 2.10 which was obtained in moderately good yield (—45%)

(Scheme 2.4) and was characterized by NMR.

KF, 180°C, 48hrs

Scheme 2.4: Reaction between 8 -aminoquinoline and 2-fluoronitrobenzene.

It was expected that reduction of the nitro group to an amine would proceed smoothly

and could be followed by ring closure and nitrogen quartemisation to form the

imidazolium salt. However, attempts to reduce the nitro group by using the same

method as for the reduction of 2 -methyl-8 -nitroquinoline, gave a complex mixture of

products. This coupled with the relative expense of the starting materials, poor yields

and time constraints led to this synthesis being abandoned.

The symmetrical quinoline imidazolium salts could not be formed according to the

established procedures (Scheme 2.5) . This was possibly due to salt formation via

the acid reacting with the quinoline function and thereby influencing solubilities. The

resulting product had extremely low solubilities in a large range of solvents including

DMSO.R R R

O n -P rO H firm M a R M T U P W n M M M n . J.n-PrOH, 60C NaBH4, THF, HCI .NH.HCI HC(OEt)+ 2 RNH2 3

0 7 NH.HCI N* k k

Scheme 2.5: Reaction scheme showing the stepwise formation of a symmetrical

imidazolium salt.


It is a matter of interest that the diamine (2.11) was also formed from the considerably

cheaper hydroxy starting material according to a literature method (Scheme 2.6).

This method was also used by T. Okada24 to make N,N-di-8-quinolyl-l,3-

propanediamine (2.12 Scheme 2.7), which, providing ring closure could be achieved,

would be a useful precursor to form the six membered NHC (2.14). Ring closure by

standard methods also failed to yield the cyclized product of 2 .1 1 .

2

OH

.NH,

NH,

Sodium Disulphate(IV) EtOH, 11 PC

1 W eek

Scheme 2.6: Synthesis of 2.11 adapted from the literature method of diamine

synthesis.

-NH

NNH H2C=0

N

N

>N

NNBS

N

Nit?

NBS= N-bromosuccinimide.

2.12 2.13 2.14

Scheme 2.7: Proposed reaction scheme to form the symmetrical 6 membered

imidazolium salt.

2.14 could be theoretically synthesised according to the preparative routes of other six

membered NHCs . Formation of 2.13 could be achieved by cyclization with aqueous

formaldehyde and the imidazolium salt obtained by reaction of 2.13 with N-

bromosuccinimide.


2.2.1.2 An Octahydroacridine based imidazolium salt

Further to the quinoline functionalised salts it was of interest to form a ligand that

would have increased hemilability as well as a chiral aspect for catalytic applications.

The aromaticity and therefore lack of sp hybridized carbons of the quinoline based

systems made them unsuitable for the task, thus octahydroacridine (Figure 2.7) was

thought to be a suitable candidate. Octahydroacridine can offer a secondary donor

function for chelation as well as sp hybridized carbons. These tetrahedral carbons can

offer the ability to make the ligand chiral as well as provide a framework for the

addition of extra steric bulk if needed.

8 1

6 3

5 4

Figure 2.7: 1,2,3,4,5,6,7,8-octahydroacridine with atom labelling.

Mannich

SOCI.

Scheme 2.8: Stepwise synthesis for 2.14.


Due to the lack of commercial availability of octahydroacridine, the established

synthetic procedures of Paine26 and Bell27 were considered. Both methods were used,

however the former was the preferable. Bell’s method proved to have fewer synthetic

steps as well as less forceful reaction conditions. However the yield stated within the

paper (50% overall) was subject to very specific reaction conditions and deviation

from the procedure led to the formation of large amounts of by products. As well as

having one extra synthetic step, Paine’s method also has the disadvantages of more

forceful reaction conditions and the overall synthetic procedure took a longer period

of time. However the procedure was found to be more reliable and gave cleaner

products. Scheme 2.8 shows Paines synthetic procedure as well as the subsequent

steps for imidazolium salt synthesis.

In order to halogenate the octahydroacridine system for subsequent N-alkylation, it

was first necessary to functionalise it with a hydroxyl group. Paine’s procedure

includes a synthetic strategy for the functionalisation with a hydroxyl group in the 4

position of the ring. This involves formation of the N-oxide by reaction with 3-

chloroperoxybenzoic acid and following work up, reaction with an excess of boiling

acetic anhydride. An alternative method however was used and adapted from the

procedure of Fontenas28, which replaces the large excess of acetic anhydride with

trifluoroacetic anhydride with the reaction being carried out at room temperature. This

method gave the desired hydroxyl substituted octahydroacridine (2.12). Conversion to

4-chlorooctahydroacridine (2.13) was achieved by reaction with thionyl chloride

which upon work up gave a stable yellow solid. 1 -mesitylimidazole was made

according to the literature method .

The final step of the reaction scheme shows the reaction between 4-

chlorooctahydroacridine (2.13) and 1-mesitylimidazole. As previously noted N-

alkylation follows a typical Sn2 pathway and, as the alkylhalide in question is a

secondary alkylchloride, we would expect the reactivity to be low. The 4-

chlorooctahydroacridine was reacted with the mesityl imidazole ( 1 :1 ) under extremely

forcing reaction conditions. The mixture was dissolved in THF and held at 90°C in an

ACE pressure tube for periods of up to 14 days. This was successful in producing a

precipitate of the desired imidazolium salt, though the reaction had to be carried out

over long periods to obtain sufficient quantities for analysis and further reactions.

Chapter Two: Chelating Imidazolium Salts 53

Crystals suitable for X-Ray diffraction were grown by layering a DCM solution of the

salt with hexanes. The crystal structure of 2.14 and a numbering scheme are given in

Figure 2.8, along with selected bond lengths and angles (Table 2.2). It should be noted

that the final imidazolium salt (2.14) was obtained as a racemate and no attempt was

made during the course of this work to separate the enantiomers.

C 7 C 6C 9

C 8C 1 1C 1 0

C 5C12,C 4

N2 ,C 1

C 1 3 C 2 3C 1 4C 3

C 2 2

C 2 1N1C 2

C 1 7

C 1 8C 2 4 C 2 0C 1 9 C 2 5

(a)

Figure 2.8: ORTEP representations of (2.14) (50% probability of thermal

ellipsoids). Hydrogen and chloride atoms omitted for clarity.


(b) w (c)

Figure 2.8 cont.: ORTEP representations of (2.14) (50% probability of thermal


Cl N1 1.328(6) C4 N2 1.469(6) N1 Cl N2 108.5(4)

Cl N2 1.334(6) C4 C16 1.516(7) Cl N1 C17 122.5(4)

C2 C3 1.361(7) C16 N3 1.349(6) Cl N2 C4 121.5(4)

C2 N1 1.382 (6 ) C17 N1 1.457(6) N3 C16 C4 118.4(4)

C3 N2 1.381(6)

Table 2.2: Bond lengths (A) and angles (°) for 2.14

Figure 2.8 shows the Cl proton is pointed towards the opposite side of the molecule

to the N3. The planes of the imidazole and the octahydroacridine form an angle of

approximately 71.8° and the planes of the imidazole and mesityl form an angle of

approximately 91.0°. This divergence from ring coplanarity is most likely due to the


influence of combined steric crowding from the mesityl’s C23 and C24 protons and

the C5 proton of the octahydroacridine function with the imidazole C2 and C3 protons.

The N1-C1-N2 of the NHC make an angle of 108.5(4)° with bond lengths of 1.328(6)

and 1.334(6) respectively and are not unusual18,19.

Due to the severe reaction conditions used in the synthesis of 2.14 an alternative

procedure was developed with the aim to form 2.15 (Scheme 2.9) which replaces the

mesityl imidazole with 1 -methylimidazole. 1 -methyl imidazole can act in this case as

both solvent and reagent and reduces the hazards of the reaction due to its high boiling

point (198°C). The reaction was carried out by the addition of 4-

chlorooctahydroacridine to an excess of 1-methylimidazole and stirred at 90°C for one

week. The resulting precipitate was analyzed by !H NMR which indicated that the

desired imidzolium salt was not formed. The *H NMR gave singlets at 9.9ppm (2H),

8.7 (2H), 7.9 (2H), 7.0 (2H) and 4.0 (6 H). The peak at 9.9 however is typical of an

imidazolium salt and so in an attempt to clarify the structure of the product, crystals

suitable for X-ray diffraction were grown. The solid state structure of the undesired

product (2.16) is shown in Figure 2.9. N-alkylation of this type usually requires the

alkyl halide to be an excess, so the desired imidazolium salt may have formed but in

extremely small quantities and lost during the work up. Compound 2.16 may have

formed by reaction of methyl imidazole with an unknown decomposition product of

the chlorooctahydroacridine, or less likely by reaction with DCM introduced to the

reaction mixture as a contaminant. Established procedures for the synthesis of 2.16

(diiodide salt) use a mixture of 1-methyl imidazole and diiodomethane (2:1) in THF,

in a pressure tube at 130°C for 16 hours . Further attempts to synthesis 2.15 were

abandoned due to time constraints.


Scheme 2.9: Reaction scheme showing the desired (2.15) and undesired (2.16)

imidazolium salts.

C 3

N 2 N1C1

C 5

C 2

Figure 2.9: ORTEP representation of (2.16) (50% probability of thermal


The crystal structure of 2.16 shows the Nl- C2 and C2- N2 bond lengths to be 1.325(5)

and 1.329(6) respectively and form an angle of 108.0(4) and are not unusual. Crystal

structures of 2.16 as a chelating ligand in complexes of Pd, Pt and Ni have been

reported in the literature30'33.

In order to extend ligand series, dicyclopentylpyridine (2.17, Figure 2.10) was formed

by substituting cyclohexanone with cyclopentanone in Paines synthetic method for


octahydroacridine. 2.17 was successfully functionalized with the hydroxyl group

(confirmed by and 13C) by the same procedure as for the octahydroacridine.

Continuing the synthetic method, halogenating the hydroxyl compound with thionyl

chloride in CH3CI caused the solution to darken. Heating the reaction mixture caused

the solution to darken further, rapidly transforming the once straw coloured solution

to a progressively blacker one. Upon work up, *H NMR gave complex spectra of

unidentified products. Time constraints again prevented further development of this

ring system in favour of pursuing the 6 membered ring system.

2.17

Figure 2.10: Dicyclopentylpyridine.

The octahydroacridine frame can offer the potential of forming a pincer ligand with

two carbenes on the 4 and 5 positions of the ring system (2.18, Figure 2.11), or a

pincer consisting of an NHC function with two octahydroacridines as N substituents

(2.19). In an attempt to form 2.18, the dihalo 4,5-dichlorooctahydroacridine was

synthesised according to the literature method (Scheme 2.10) and was reacted with

1-mesitylimidazole under the same reaction conditions as for 2.14. The formation of

2.18 would prove extremely interesting in comparison to Gibson’s pincer chromium

complex (2 .2 0 ) 34 as an active catalyst for the oligomerization of ethylenes. 2.18 was

not formed however as it was found that monosalt formation by reaction of one of the

halides rendered the compound sufficiently insoluble in THF thereby preventing

further reaction. This coupled with the length of the reaction time meant that only the

monosalt was formed. Changing the halide by using thionyl bromide in the synthetic

stage and/or improving the solubility with solvent mixes or DMF/HMPA may be

ways of overcoming this. These solutions however were not tried in this work, though

the addition of sodium iodide to the reaction mixture was tried with no observable

effects.


N

2.18

- N ^ N

N NN-

2.19

; Cr' ci ci cl

2.20

N\

Figure 2.11: Proposed imidazolium salts for ‘pincer’ ligands (2.18 & 2.19) and

Gibson’s Cr ‘pincer’ complex (2.20)

N'I

O

NCl Cl

(AcOTO

145-150C

SOCL

CO„H

Cl

N'CH3CI

N'OAc

(AcO)2Q

OAc

145-150C

N

hci/h2o

80C | ■' | 100COH OH OAc

NOAc

Scheme 2.10: Synthetic method for 4,5-dichlorooctahydroacridine.

2.2.2 Bis imidazolium salts.

Like the monofunctionalised imidazolium salts, bis imidazolium salts have generally

been synthesised by simple starting materials and involve two imidazoles separated by

a linker group, usually (CFDn with variable steric bulk on the second imidazolium

nitrogens. The simplest example of a bis-imidazolium salt would be 3,3-dimethyl-1,1-


methylenediimidazolium iodide which was used to form a palladium(II) complex

(2.21) by Herrmann et al via reaction with palladium acetate.35

2.21

Figure 2.12: Herrmann’s Pd(II) complex of 3,3-dimethyl-1,1-

methylenediimidazolium iodide.

Prior to the start of this work only a few examples of bis carbenes with chiral

backbones were reported in the literature, the first example of which was made by

RajanBabu et al (2.22, Figure 2.13). During the course of this work a similar ligand

to our own (2.27, Figure 2.14) was also described (2.23) and complexed with Pd by

Harrison et al37 (submitted 18/03/04) Our ligand however was submitted (23/02/04)

for publication in a joint paper with M. Beller and others in a study of the Pd

catalysed telomerization reaction and was published around the same time. During the

course of this work, other notable chiral bis imidazolium salts and their carbene metal

complexes were also synthesised by various other groups (2.24)39 and (2.25)40.


2Br-

2.22

2Br-

2.23

2 1 -

R 2Br" R

2.24 2.25

Figure 2.13: Examples of chiral bis imidazolium salts in the literature (at

writing)36,37,39’40.

2.2.2.1 Preparation o f an imidazolium salt DIOP analogue.

DIOP (2.26, Figure 2.14) is one of the earliest and most extensively explored chiral

phosphine ligands in asymmetric hydrogenation and has shown synthetic chemists

that the chirality does not have to be on the phosphorous centres for a phosphine

compound to be a highly selective ligand for asymmetric catalysis, while the chiral

backbone can play a significant role in the asymmetric induction41. For these reasons

a carbene analogue of DIOP (2.27) was thought to be useful both as a comparison to a

well studied phosphine ligand as well as a welcome addition to the canon of

phosphine DIOP analogues42.


o

o

PPh,

PPh,

NVI^N

N ^ N2X-

2.26 2.27

Figure 2.14: DIOP (2.26) and our target imidazolium salt analogue (2.27).

It was decided to add the imidazolium functions on to an existing chiral backbone for

synthetic simplicity. Just as RajanBabu used a binapthelene template to build upon for

2.22, we used diethyl tartrate. The synthetic work for the bis-imidazolium salt was

based largely on Brown’s improved preparation of DIOP43. The chiral backbone was

built according to Brown’s literature method43 up to step 3 (Scheme 2.11).

>OMeTos-OH r n n r H

O M e x ^ O ^ X C O O C 2H 5 LiAIH4

HO ''C 0 0 C 2H5 -M eOH q ""CooC 2H5Step 1 Step 2

Tos-CI/Pyridine v OTos 1. Methyl imidazole

\O

O0 ^ '- " w _ OTos 2. (excess) KPF6

Step 3 s tep 4

Scheme 2.11: Stepwise formation of 2.27, taken from Browns DIOP synthesis.

2.27 was smoothly transformed from the tosyl dioxolan by a two step process, first by

reaction with neat 1-methylimidazole at 90°C overnight, followed by the addition of

H2O and an excess of KPF6 to precipitate a white solid of 2.27 as the trans product.

Recrystallisation from a minimum amount of acetone gave crystals suitable for1 ilcharacterisation by X-ray diffraction. 2.27 was also characterized by H and C NMR.


The structure and numbering scheme of 2.27 are given in Figure 2.15 along with

selected bond lengths and angles in Table 2.3.

The imidazolium moieties of 2.27 are orientated about their pendant bonds so that the

C2 and C ll protons are facing towards opposite directions of the molecule, though

free rotation about the N2-C5 and N4-C8 bonds should allow for chelation to a metal

centre. The imidzolium N-C-N bond lengths and angles are typical of other

imidazolium salts.

2.27 clearly represents a precursor to the carbene analogue of DIOP, though the nature

of the carbenes means that upon chelation, 2.27 would form a nine membered chelate

ring compared to the seven membered ring of a DIOP chelate. The ligand is very rigid

and because of this would probably only form the cis product upon metal

complexation.

C 1 301

02

C 6C 4

C 1 1 C 8N 2

C 7C 1 2 C 3C 5

N 4

N 3N1

C 2<2 r ^ c 9

C 1 0C 1

Figure 2.15: ORTEP representation of (2.27) (50% probability of thermal

ellipsoids). Hydrogen and PF6 atoms omitted for clarity.


C2 N1 1.331(5) C ll N3 1.323(5) N1 C2 N2 108.8(4)C2 N2 1.328(5) C ll N4 1.327(5) Cl N1 C2 124.9(4)C3 N1 1.375(5) C12 N3 1.466(3) C2 N2 C5 124.3(5)C3 C4 1.351(5) C9 N4 1.364(3) N3 C ll N4 108.4(5)C4 N2 1.386(3) CIO N3 1.364(3) C12 N3 C ll 125.9(5)N2 Cl 1.470(6) C9 CIO 1.327(5) C8 N4 C9 125.4(3)N2 C5 1.468(6) N4 C8 1.466(5)

Table 2.3: Selected bond lengths (A) and angles (°) for (2.27)

2.3 Conclusions

A range of imidazolium salts have been synthesised and characterised as precursors to

the corresponding NHC ligands. The majority of these have combined a functional

group with an imidazolium moiety and display a range of varying degrees of steric

bulk. The quinoline based ligands offer a rigid chelate based around a delocalised

electron ring system, whilst the octahydroacridine based system offers a more flexible

chelate as well as a chiral centre. A chiral bis-imidazolium salt DIOP analogue has

also been synthesised offering the potential for comparison in enantioselective

catalysis with the well studied DIOP ligand. All the imidazolium salts were

characterised by spectroscopic means including !H, 13C NMR and Mass Spec. A few

examples have been crystalographically characterised.

2.4. Experimental

2.4.1 General comments

Unless otherwise stated all manipulations were carried out using standard Schlenk

techniques, under an atmosphere of dry argon or in a nitrogen glove box. Glassware

was dried overnight in an oven at 120°C or flame dried prior to use. Tetrahydrofuran

(THF), diethyl ether (Et2 0 ), and hexane were dried and degassed by refluxing under

dinitrogen and over sodium wire and benzophenone. Dichloromethane (DCM),

methanol (MeOH), and acetonitrile (MeCN) were dried over calcium hydride. All


other anhydrous solvents were obtained by distillation from the appropriate drying

agents under dinitrogen. Deoxygenation of solvents and reagents was carried out by

freeze thaw degassing

All NMR solvents were purchased from Aldrich and Goss, dried over 3A molecular

sieves and ffeeze-thaw degassed three times. All reagents were purchased from

commercial sources and used without purification, unless otherwise stated.

i nAll NMR data are quoted in 8 /ppm. H and C spectra were recorded on a Bruker 400

MHz DPX Avance, unless otherwise stated, and reference to SiMe4 . Electrospray

mass spectrometry (ESMS) was performed on a VG Fisons Platform II instrument by

the Department of Chemistry, Cardiff University

The following compounds were prepared by established literature methods. 8 -

imidazol-l-yl-quinoline (2.4) , octahydroacridine N-oxide , 4,5-

dichlorooctahydroacridine , mesityl imidazole, and /ra«s-4,5-bis[tosyloxymethyl]-

2,2-dimethyl-1,3-dioxolan43.

2.4.1.1 Synthesis Of Functionalised Imidazolium salts

Synthesis of 2-methyl-quinolin-8-amme (2.3):

NH2

2-methyl-8-nitroquinoline (1.09g) in 30ml THF was placed in a lit flask with a

catalytic amount of Pd on Carbon (10% mol) and a drop of MeOH. The atmosphere

was then replaced with H2 and the mixture allowed to stir overnight. The solution was

then filtered through celite and the solvent removed to give a pale yellow powder. *H

NMR (CDCI3 , 400MHz, 8 ) 7.9 (m, 1H, Quin H), 7.15 (m, 2H, Quin H), 7.05 (m, 1H,

Quin H), 6.85 (m, 1H, Quin H), 4.9 (s, br, 2H, NH2), 2.60 (s, 3H, CH3).

C h a p t e r T w o : C h e l a t i n g I m i d a z o l i u m S a l t s

Synthesis of 8-ImidazoI-I-yl-quinoline (2.4).

8 - I m i d a z o l - l - y l - q u i n o l i n e w a s m a d e a c c o r d i n g t o a m o d i f i e d l i t e r a t u r e m e t h o d 1 4 . 8 -

A m i n o q u i n o l i n e ( 5 g , 3 5 m m o l ) i n 1 2 0 m l M e O H w a s m i x e d w i t h 4 0 % g l y o x a l ( 5 . 0 7 g ,

3 5 m m o l ) o v e r n i g h t t o g i v e a y e l l o w m i x t u r e . N H 4 C I ( 3 . 7 g , 6 9 m m o l ) a n d 3 7 % a q .

f o r m a l d e h y d e ( 5 . 6 7 g , 6 9 m m o l ) w a s a d d e d t o t h e m i x t u r e w h i c h w a s t h e n d i l u t e d w i t h

M e O H ( 1 0 0 m l ) a n d r e f l u x e d f o r o n e h o u r . H 3 P O 4 ( 5 . 5 m l , 8 5 % ) w a s t h e n s l o w l y

a d d e d t o t h e m i x t u r e b e f o r e b e i n g r e f l u x e d o v e r n i g h t . A f t e r r e m o v a l o f t h e s o l v e n t t h e

d a r k r e s i d u e w a s p o u r e d o n t o i c e ( 4 0 0 g ) a n d n e u t r a l i z e d w i t h a q . 4 0 % K O H u n t i l p H 9 .

T h e r e s u l t i n g m i x t u r e w a s t h e n e x t r a c t e d w i t h E t 2 0 ( 4 x 1 5 0 m l ) . T h e e t h e r e a l p h a s e

w a s w a s h e d w i t h H 2 O , b r i n e a n d d r i e d w i t h N a 2 S 0 4 . T h e o r g a n i c s o l v e n t w a s

r e m o v e d t o g i v e a b r o w n s o l i d o f 8 - i m i d z o l - l - y l - q u i n o l i n e ( 1 . 7 7 g , 2 6 % ) w h i c h w a s

u s e d w i t h o u t f u r t h e r p u r i f i c a t i o n . 1H N M R ( M e O D 2 5 0 M H z , 8 ) : 7 . 2 8 ( s , 1 H , N C H N ) ,

6 . 7 8 ( m , 1 H , A r - H ) , 6 . 6 1 ( m , 1 H , A r - H ) , 6 . 3 6 ( m , 1 H , A r - H ) , 6 . 1 9 ( m , 1 H , A r - H ) 6 . 0 6

( m , 1 H , A r - H ) , 5 . 9 9 ( s , 1 H , A r - H ) , 5 . 9 6 ( m , 1 H , A r - H ) 5 . 5 9 ( s , 1 H , A r - H ) . 1 3 C N M R

( C D C I 3 , 1 0 0 M H z , 8 ) : 1 5 1 . 1 , 1 4 1 . 8 , 1 3 8 . 9 , 1 3 6 . 8 , 1 3 4 . 3 , 1 2 9 . 7 , 1 2 8 . 4 , 1 2 7 . 6 , 1 2 6 . 4 ,

1 2 7 . 9 , 1 2 2 . 3 , 1 2 1 . 7 .

Synthesis of 8-(lH-imidazol-l-yl)-2-methylquinoline (2.5):

N

8 - ( l H - i m i d a z o l - l - y l ) - 2 - m e t h y l q u i n o l i n e w a s m a d e u s i n g t h e s a m e p r o c e d u r e a s f o r

2.4. 2 - m e t h y l - q u i n o l i n - 8 - a m i n e ( 1 . 4 0 g , 8 . 8 m m o l ) w a s m i x e d w i t h 4 0 % g l y o x a l

( 1 . 2 7 g , 8 . 8 m m o l ) i n 3 0 m l M e O H o v e r n i g h t t o g i v e a y e l l o w m i x t u r e . NH4 CI ( 0 . 9 4 g ,

1 7 . 6 m m o l ) a n d 3 7 % A q . f o r m a l d e h y d e ( 1 . 4 2 g , 1 7 . 6 m m o l ) w e r e t h e n a d d e d t o t h e

m i x t u r e w h i c h w a s d i l u t e d b y t h e a d d i t i o n M e O H ( 1 5 0 m l ) . T h e m i x t u r e w a s r e f l u x e d

f o r a n h o u r . H 3 P O 4 ( 1 . 6 m l , 8 5 % ) w a s t h e n s l o w l y a d d e d t o t h e m i x t u r e b e f o r e b e i n g


refluxed overnight. After removal of the solvent the dark residue was poured onto ice

(lOOg) and neutralized with aq. 40% KOH until pH9. The resulting mixture was then

extracted with Et2 0 (4x100ml). The ethereal phase was washed with H2O, brine and

dried with Na2 SC>4 . The organic solvent was removed to give the light brown 8 -(lH-

imidazol- 1 -yl)-2-methylquinoline. (0.56g, 31%). *H NMR(CDC13 250MHz, 8 ): 8.15

(s, 1H, NCHN), 8.00 (m, 1H, Quin H4), 7.70 (m, 1H, Quin H), 7.55 (m, 1H, Quin H),

7.45 (m, 2H, Quin H) 7.25 (m, 1H, HCCH), 7.15 (m, 1H, HCCH), 2.60 (s, 3H, CH3).

UC NMR (CDC13, 100MHz, 8 ): 159.0, 140.5, 138.5, 136.0, 135.5, 128.5, 127.0,

126.8,125.0,123.0,122.5, 120.5,25.0

Synthesis of l-methyl-3-quinoline-imidazolium iodide (2.6):

Methyl iodide (2g) was added to a solution of 8 -imidazol-l-yl-quinoline (0.52 lg,

3.3mmol) in 20ml THF. The resulting mixture was allowed to stir overnight at room

temperature. The resulting yellow/orange precipitate was filtered using a filter stick

and washed with further amounts of fresh THF (3x5ml) to afford a brown powder.

This was then recrystalized from DCM/Hexane to give the desired imidazolium salt.

(0.86g, 87%) ‘H NMR(CDC13, 400MHz, 8 ): 10.2 (s, 1H, NCHN), 8.95 (m, 1H,

quinH2), 8.30 (m, 2H, HCCH), 8.00 (m, 1H, Quin H4), 7.85 (m, 1H, Quin H6 ), 7.70

(m, 1H, Quin H8 ), 7.50-7.6 (m, 2H, Quin H3,7) 4.3 (s, 3H, NCH3) MS (ES) m/z

(%): 210.2 (38) [M-I]+, 155.0 (59), 141.0 (8 8 ), 139.9 (100).

Synthesis of l-benzyl-3-quinolinimidazolium bromide (2.7):


Benzyl bromide (2g) was added to a solution of 8 -imidazol-l-yl-quinoline (1.03g,

6 .8 mmol) in 20ml THF. The resulting mixture was allowed to stir overnight at room

temperature. The resulting yellow/brown precipitate was filtered using a filter stick

and washed with fresh THF (3x5ml) to afford a brown powder. This was then

recrystallized from DCM/Hexane to give the desired imidazolium salt (1.7g, 75%). rH

NMR(C6D6, 250MHz, 5): 10.75 (s, 1H, NCHN), 8.85 (m, 1H, Quin H2), 8.20 (m, 2H,

Ar H), 7.95 (m,2H, Ar H, HCCH), 7.60-7.75 (m, 4H, HCCH Ar H), 7.50 (m, 1H, Ar

H), 7.35 (m, 3H, Ar H), 5.85(s, 2H, CH2-Ph). MS (ES) m/z (%): 285.3 (73) [M-Br]+,

155.0 (48), 127.9 (59), 114.8 (100). MS (ESI) m/z (%): found: 285.1449 [M-Br]+;

expected: 258.3482

Synthesis of l-methyl-3-(2-methyl-quinoline)-imidazolium iodide (2.8):

i-

Methyl iodide (2g) was added to a solution of 8-(lH-imidazol-l-yl)-2-

methylquinoline (lg 4.67mmol) in 20ml THF. The resulting mixture was allowed to

stir overnight at room temperature. The resulting yellow/orange precipitate was

filtered using a filter stick and washed with THF (3x5ml) to afford a brown powder.

This was then recrystallized from DCM/Hexane to give the desired imidazolium salt

(1.4g, 84%). *H NMR (CDC13, 400MHz, 8 ): 10.15 (s, 1H, NCHN), 8.15 (m, 2H,

HCCH), 7.8-7.95 (m, 2H, Quin H4,6), 7.55-7.7 (m, 2H, Quin H7,8), 7.35 (m, 1H,

Quin H3), 4.30 (s, 3H, NCH3), 2.65 (s, 3H, CH3). ,3C NMR (CDC13, 100MHz, 8 ):

161.30, 139.9, 137.4, 136.7, 130.4, 130.3, 127.5, 125.9, 125.3, 124.4, 123.9, 123.3,

37.7,25.8. MS (ES) m/z (%): 224.2 (35) [M-IJ+, 181.1 (52), 154 (95), 139.9 (100).


Synthesis of l-benzyl-3-(2-methyl-quinoline)-imidazolium bromide (2.9):

Benzyl bromide (2g) was added to a solution of 8-(lH-imidazol-l-yl)-2-

methylquinoline (1.03g 4.93mmol) in 20ml THF. The resulting mixture was allowed

to stir overnight at room temperature. The resulting yellow/orange precipitate was

filtered using a filter stick and washed with THF (3x5ml) to afford a brown powder.

This was then recrystallized from DCM/Hexane to give the desired imidazolium salt

(1.2g, 63%). ‘H NMR (CDC13, 250MHz, 8): 10.7 (s, 1H, NCHN), 8.15 (m, 2H, ArH),

7.85 (m, 2H, ArH), 7.45-7.65 (m, 3H, ArH) 7.35 (m, 5H, ArH), 5.85 (s, 2H, CH2-Ph),

2.65(s, 3H, Quin CH3). MS (ES) m/z (%): 300.3 (98) [M-Br]+, 286.3 (18), 181.1 (19),

155.0(25), 114.8 (100).

Synthesis of N, N’ - di-8-quinolyl-l,2-ethanediamine (2.11):

An aqueous mixture of 8 -hydroxyquinoline (2.90g, 0.02mol), 1,2-ethanediamine (0.4g,

0.01 mol) and sodium disulphate (3.80g, 0.02mol) was refluxed for one week at 110°C.

Recrystallization of the subsequent orange solid from ethanol gave light orange plates

of 2.11 (0.6g, 18%). 'H NMR (CDCI3 , 400MHz, 8 ): 8 . 6 (m, 2H), 7.95 (m, 2H), 7.3

(m, 4H), 7.0 (m, 2H), 6.65 (m, 2H), 6.3 (s, br, 2H, NH), 3.65 (s, 4H).


Synthesis of Octahydroacridine N-oxide:

O

Octahydroacridine N-oxide was made according to the stepwise literature method of

Paine25.

Synthesis of l,23?4,5,6,7,8-octahydroacridin-4-ol (2.12): (Adapted from the-no

literature method )

OH

Trifluoroacetic anhydride (10.2ml, 0.072mol, 2.5 equivalents) was slowly added to a

stirred solution of octahydroacridine N-oxide (5.85g, 0.0288mol) in dry DCM (50ml)

causing a slight increase in temperature of the solution. The solution was allowed to

stir for a further hour at room temperature. The volatiles were then removed under

reduced pressure to leave a yellow viscous residue, which was then taken up in 50ml

DCM. This was then saponified by the addition of a 2M solution of sodium carbonate,

the biphasic mixture being vigorously stirred for 3 hours. The organic phase was then

separated and the aqueous phase washed twice with DCM. The combined organic

extracts were then combined and washed with water and brine and then dried over

magnesium sulphate. The solvent was then removed to give octahydroacridin-4-ol

(4.98g, 85%). 'H NMR (CDCI3 , 250MHz, 8 ): 7.05 (s, 1H, ArH), 4.55 (m, 1H,

CtfOH), 4.05 (Br s, 1H, CHOtf), 2.8 (m, 2H, CH2), 2.7 (m, 4H, CH2), 2.2 (m, 1H,

CH2), 1.6-2.0 (m, 7H, CH2). ,3C NMR(CDC13, 100MHz, 8 ): 154.86, 154.35, 137.5,

130.97, 128.55,68.36,31.88, 31.15,28.42,27.99,23.16,22.79, 19.39.

Synthesis of 4-chlorooctahydroacridine (2.13): (Adapted from the literature method)


Thionyl chloride (30ml) in CHCI3 (30ml) was added to a solution of

octahydroacridin-4-ol (14g, 0.069mol) in 50ml DCM. The mixture was stirred at

room temperature for 10 minutes, then refluxed for an hour at 80°C. The excess

thionyl chloride and volatile by-products were then removed under reduced pressure

and the residue dissolved in DCM (150ml). The solution was then extracted with

Na2CC>3 (15g) in water (250ml) and the aqueous phase (pH= 8.5) was washed with

DCM (2x200ml). The DCM extract was then dried with Na2CC>3 and the solvent

removed under reduced pressure to give a yellow solid of 4-chlorooctahydroacridine

(9.80g, 64.3%). 'H NMR (CDC13, 250 MHz, 6 ): 7.05 (s, 1H, ArH), 5.20 (m, 1H,

C//C1), 2.55-2.95 (m, 6 H, CH2), 2.05-2.35 (m, 3H, CH2), 1.65-1.85 (m, 5H, CH2). 13C

NMR (CDCI3 , 100MHz, 6 ): 155.53, 151.22, 137.96, 132.35, 128.97, 59.44, 32.65,

32.24,28.6,27.54,23.15,22.65, 17.42.

Synthesis of 4,5-dichlorooctahydroacridine:

Cl Cl

4,5-dichlorooctahydroacridine was made according to the literature method of Gan26.

Synthesis of l-mesityl-3-octahydroacridinimidazolium chloride (2.14):

Cl-

4-chlorooctahydroacridine (lg, 4.5mmol) and 1- mesityl imidazole (0.84g 4.5mmol)

were placed in an ACE pressure tube with 10ml THF. The mixture was refluxed at

90°C for 10 days as a dark precipitate formed. The precipitate was filtered and washed

with Et2<3 (4x10ml) to give a beige solid of the imidazolium salt (0.2g, 11%). The

filtrate was placed back under reflux where it continued to react further over time

(13% overall). *H NMR (CDC13, 400MHz, 5) 10.15 (s, 1H, NCHN), 8.18 (s, 1H,


HCCH), 7.65 (s, 1H, HCCH), 7.20 (s, 1H, ArH), 6.95 (m, 2H, ArH), 6.55 (s, 1H,

CHN), 1.80-3.40 (m, 23H, CH2, CH3) MS (ES) m/z (%): 372.3 (29) [M-C1]+, 186.2

(70), 170.1 (97), 156.0 (100), 142.0 (97), 114.8 (96). MS (ESI) m/z (%): found

372.3424 [M-C1]+; expected: 372.5333.

Synthesis of l,l-dimethyl-3,3-methylene-diimidazolium dichloride (2.16):

2.16 was synthesised during the attempted synthesis of 2.15. 2.13 (lg, 4.5mmol) was

dissolved in 1-methylimidazole (5g, 60mmol) and stirred at 90°C for one week. The

resulting beige precipitate was filtered and recrystallized from DCM to give 2.16

(0.4g). 'H NMR (CDCI3 ,400MHz, 8 ): 9.9ppm (s, 2H), 8.7 (s, 2H), 7.9 (s, 2H), 7.0 (s,

2H) and 4.0 (s, 6 H).

Synthesis of Dicyclopentylpyridine (2.17):

2.17 was made according to the modified stepwise procedure of Paine25.

2.4.1.2 Synthesis of Bis Imidazolium SaltsSynthesis of (-)-/ra/f s-4,5-bis [tosyloxymethyl]-2,2-dimethyl-1,3-dioxolan:

OTos

— OTos

(-)-^r<my-4,5-bis[tosyloxymethyl]-2,2-dimethyl-1,3-dioxolan was made according to a

literature method43.

Synthesis of (-)/ra«s-4,5-Bis(methylimidazolium hexaflurophosphate)-2,2-

dimethyl-13-dioxolan (2.27):


(-)-/raw.s-4,5-bis[tosyloxymethyl]-2,2-dimethyl-1,3-dioxolan (5g, 10.6mmol) was

stirred with 1-methylimidazole (15ml) overnight at 90°C to give a brown/red solution.

100ml of water was then added followed by an excess of KPF6 (4.57g) to give a white

precipitate. This was then filtered and recrystalized from acetone and Et2 0 to give

white crystals of the bis-imidazolium salt (5.2g, 84%). Crystals suitable for X-ray

determination were grown by the slow evaporation of an acetone solution. NMR

(CD3CN, 400MHz, 5): 8.3 (s, 2H, NCHN), 7.2-7.25 (m, 4H, HCCH), 4.3 (m, 2H,

OCH), 4.1 (m, 2H, CH2), 3.9 (m, 2H, CH2), 3.65 (s, 6 H, NCH3), 1.15 (s, 6 H, CCH3).

13C NMR (CD3CN, 250MHz, 8 ): 136.4 (NCN), 123.5 (NCCN), 123.2 (NCCN), 111.1

(OCC2H6) 75.2 (OCH), 50 (CH2), 35.7 (NCH3), 25.8 (CH3). MS (ES) m/z (%): 437.2

(82) [M-PF6]+, 281.3 (24), 151.0 ( 1 0 0 ).

2.5 References:

1. H. Wanzlick, H. Kleiner, Angew. Chem., 1961, 79, 493.

2. J. Teles, J. Melder, K. Ebel, R. Schneider, E. Gehrer, W. Harder, S. Brode, D.

Enders, K. Breuer, G. Raabe, Helv. Chim. Acta, 1996, 79, 61-83.

3. D. Ender, K. Breuer, G. Raabe, J. Runsink, J. Teles, J. Melder, K. Ebel, S.

Brode, Angew. Chem. Int. Ed., 1995. 34, 9, 1021-1023.

4. N. Kuhn, T. Kratz, Synthesis, 1993, 561.

5. W. A. Herrmann and C. Kocher, Angew. Chem. Int. Ed. Engl., 1997, 36, 2162-

2187.

6 . K. J. Cavell, D. S. McGuinness, Coordin. Chem. Rev., 2004,248, 671-681

7. D. S. McGuinness, K. J. Cavell, Organometallics, 2000,19, 741-748.

8 . A. Tulloch, A. Danopoulos, R. Tooze, S. Cafferkey, S. Kleinhenz, M.

Hursthouse, Chem., Commun., 2000, 1247-1248


Comm., 2004, 1612-1613


10. K. Coleman, H. Chamberlayne, S. Turberville, M. Green, A. Cowley. J. Chem.

Soc. Dalton Trans., 2003, 2917-2922.

11. M. Froseth, A. Dhindsa, H. Roise, M. Tilset, J. Chem. Soc. Dalton Trans.,

2003, 4516-4524.

12. R. Douthwaite, J. Houghton, B. Kariuki, Chem. Comm., 2004, 698-699.

13. P. Arnold, S. Mungur, A. Blake, C. Wilson, Angew. Chem. Int. Ed., 2003,

5981-5984.

14. H. Lee, J. Zeng, C. Hu, M. Lee, Inorg. Chem., 2004, 43, 6822-6829

15. J. Liu, J. Chen, J. Zhao, Y. Zhao, L. Zhang, Synthesis, 2003,17, 2661-2666.

16. A. Kiyomori, J. Marcoux, S. Buchwald, Tetrahedron. Lett., 1999, 40, 2657-

2660

17. M. Gardiner, W. Herrmann, C. Reisinger, J. Schwarz, M. Spiegler, J.

Organomet. Chem., 1999, 572, 239-247.

18. A. Danopoulos, S. Winston, T. Gelbrich, M. Hursthouse, R. Tooze, Chem.

Comm., 2002, 482-483.

19. D. Nielsen, K. Cavell, B. Skelton, A. White, Inorg., Chim., Acta, 2002, 327,

116-125

20. J. J. Kulagowski, C. W. Rees, Synth. Comm., 1980, 03, 215.

21. M. Zettlizer, T. H. Dieck, E. Haupt, L. Stamp, Chem Ber. 1986, 119, 1868-

1875

22. A. J. Arduengo, R. Krafczyk, R. Schmutzler, Tetrahedron, 1999, 55, 14523-

14534

23. K.A Jensen, P.H. Nielsen, Acta Chem. Scand, 1966, 20, 1113

24. T. Okada, M. Yoshida, T. Hirose, K. Kasuga, T. Yu, M. Yuasa, I. Sekine,

Electrochimica Acta, 2000, 45, 4419-4428

25. M. Mayr, K. Wurst, K. Ongania, M. Buchmeiser, Chem., Eur., J., 2004, 10,

1256-1266.

26. X. Gan, S. Parvenn, W. L. Smith, E. N. Duesler, R. T. Paine, Inorg. Chem.,

2000, 39, 4591-4598.

27. M. L. Pilato, V. J. Catalano, T. W. Bell, J. Org. Chem., 2001, 6 6 , 1525-1527.

28. C. Fontenas, E. Bejan, H. Haddou, G. G. A. Balavoine, Synth. Comm., 1995,

25 (5), 629-633.

29. A. J. Arduengo, F. P. Gentry, P. K. Taverkere, H. E. Simmons, Du pont de

Nemours and company, US 6177575 2001.


30. M. Muehlhofer, T. Strassner, E. Herdtweck, W. Herrmann, J. Organomet.

Chem., 2002, 660, 121-126

31. W. Fehlhammer, T. Bliss, U. Kembach, I. Brudgam, J. Organomet. Chem.,

1995, 490, 149-153.

32. W. Herrmann, C. Reisinger, M. Spiegler, J. Organomet. Chem., 1998, 557,

93-96.

33. W. Herrmann, J. Schwarz, M. Gardiner, M. Spiegler, J. Organomet. Chem.,

1999, 575, 80-86.

34. D. S. McGuinness, V. C. Gibson, D. F. Wass, J. W. Steed, J. Am. Chem. Soc.,

2003,125, 12716-12717

35. W. A. Herrmann, M. Elison, J. Fisher, C. Kocher, G. Artus, Angew. Int. Ed.

Engl., 1995, 34, No 21

36. D. S. Clyne, J. Jin, E. Genest, J. C Gallucci, T. V. RajanBabu, Org. Lett. 2000,

2, 8 , 1125-1128

37. C. Marshall, M. F. Ward, W. T. A. Harrison, Tet. Letters, 2004, 45, 5703-5706.

38. R. Jackstell, S. Harkal, H. Jiao, A. Spannenberg, C. Borgmann, D. Rottger, F.

Nierlich, M. Elliot, S. Niven, K. Cavell, O. Navarro, M. S. Viciu, S. P. Nolan,

M. Beller, Chem. A Europ. Jour., 2004,10, 16, 3891-3900.

39. W. Duan, M. Shi, G. Rong, Chem. Comm., 2003, 2916-2917

40. L. Bonnet, R. Douthwaite, R. Hodgson, Organometallics, Comm., 2003, 22(22)

4384-4386

41. D. Liu, W. Li, X. Zhang, Tetrahedron: Asymmetry, 2004,15, 2181-2184.

42. (a) G. Zhu, P. Cao, Q. Jiang, X. Zhang, X. J. Am. Chem. Soc., 1997,119, 1799;

(b) G. Zhu, X. Zhang, J. Org. Chem., 1998, 63, 9590; (c) G. Zhu, A.

Casalnuovo, X. Zhang, J. Org. Chem. 1998, 63, 8100; (d) G. Zhu, X. Zhang,

Tetrahedron: Asymmetry, 1998, 9, 2415; (e) G. Zhu, Z. Chen, X. Zhang, J. Org.

Chem. 1999, 64, 6907; (f) P. Cao, X. Zhang, J. Org. Chem. 1999, 64, 2127; (g)

S. Lee, Y. Zhang, C. Song, J. Lee, H. Choi, Angew. Chem., Int. Ed., 2002, 41,

847; (h) S. Lee, Y. Zhang, Org. Lett. 2002, 4, 2429; (i) S. Lee, Y. Zhang,

Tetrahedron: Asymmetry, 2002,13, 1039; (j) W. Li, J. Waldkirch, X. Zhang, J.

Org. Chem., 2002, 67, 7618.

43. B. A. Murrer, J. M. Brown, P. A. Chaloner, P. N. Nicholson, D. Parker,

Synthesis Comm. 1979, 350- 352.

Chapter Three: Ag(I) Complexes Of Chelating Carbenes

Chapter Three

Silver (I) Complexes Of Chelating Carbene

Ligands

3.1 Introduction.

3.1.1 Ag(I)(NHC) As A Transmetallation Agent

Due to the catalytic importance of palladium NHC complexes, and more importantly

Pd(hydrocarbyl)(NHC), the main aim of this work is the synthesis of novel complexes

of these types. The last few years however have shown the importance of Ag(I)(NHC)

complexes as intermediates in the synthesis of their related Pd complexes. In

particular, Ag intermediates have been shown to be extremely useful when the desired

complex contains a functionalised carbene ligand1. It is therefore appropriate to

discuss some of the history behind Ag(NHC)s and why they can play such a pivotal

role in Pd(NHC) synthesis.

Early synthetic procedures for Ag carbenes followed the standard routes as for any

other metals at that time, the first being reported by Arduengo2 in 1993. Arduengo

displaced the weakly coordinated triflate ligand from Ag(OTf) with 1,3-

dimethylimidazolin-2-ylidine and formed the homoleptic complex (3.1). (Figure 3.1)

Mes MesI I

Mes Mes3.1

Mes = Mesityl

Figure 3.1: Arduengo’s homoleptic Ag complex.


In the late 1990’s Lin and Wang were conducting work on the molecular recognition

of coinage metals and were investigating possible ligand-unsupported Agl-AgI

interactions in Ag-carbene compounds. They published a paper3 that demonstrated the

ability of these Ag(NHC)s to act as transmetallation agents, and in doing so, offering

a potentially easily accessible route to form palladium complexes. In particular early

reports have since shown this to be a valuable and easy synthetic route to

Pd(II)(hydrocarbyl)functionalized carbene complexes. The report concerned the

synthesis of Ag(I) complexes of l,3-diethylbenzimidazol-2-ylidene and the

subsequent transfer of the carbene ligand onto Pd and Au. This was achieved by

stirring the imidazolium salt with Ag2 0 in DCM, or with AgBr and NaOH under

phase transfer conditions, which afforded the Ag complex (3.2) (Figure 3.2) in high

yields. Ligand transfer was achieved simply by stirring the Ag complexes with

Pd(MeCN)2Cl2 to give 3.3 (Figure 3.2) or with Au(SMe2)Cl to give 3.4 (Figure 3.2).

The silver halide by-products were shown to be recycled back in to the Ag(NHC) by

phase transfer conditions. Importantly the silver complexes were shown to be stable

towards air and, as water is a by-product of the reaction, the complexes were also

stable towards moisture.

EtI I

)—9_(I A a Ni r ¥ i

Et Br Br Et

3.2Au-Br

3.3

3.4

Figure 3.2: Lin and Wang’s Ag complex (3.2) and its related transmetallation

products.

Realising this method’s potential for the use in group 10 metal carbene complex

synthesis, several groups have employed it as a reliable synthetic strategy1,4’12.

Amongst these groups, Cavell successfully synthesised hydrocarbylpalladium


complexes with functionalised carbenes such as 3.51 (Figure 3.3). The traditional

routes to synthesise this type of ligand would have required strong bases with which

to deprotonate the imidazole C2-H for the formation of the free carbene, often leading

to abstraction of acidic (or ‘active’) protons on the CH2 linker. In the case of 3.5, both

deprotonation to form the free carbene as well as reaction with Pd(OAc)2 as a direct

route to the Pd complex was unsuccessful. However, by simply reacting the Ag

complex with PdMeCl(COD), the desired Pd complex was synthesised.

MePd

3.5

Figure 3.3: Cavell’s hydrocarbylpalladium complex.

Further to the successful application of a Ag method with functionalised mono-

imidazolium salts, the same group also implemented the Ag route for the formation of

functionalised pincer complexes5. A one-pot reaction method was also developed that

produced high yields of the desired Pd(II) complex. (Scheme 3.1) 9

/

2Br-

a. Ag20 , b. AgBF4, c. PdCIXL2

DMSO, rt.Pd

X= Cl; l_2= MeCN

BF.

X= Me; l_2= COD

Scheme 3.1: Cavell’s ‘one pot’ synthesis of Pd hydrocarbyl species via the Ag

route.

The relative ease of formation of Ag(I) functionalised NHC’s and their valuable

ability to act as transmetallation agents, in particular for the formation of hydrocarbyl

species, led to the method being adopted as the route for which to form the Pd

complexes within this work.


3.1.2 Structural characteristics of Ag(I) Carbene complexes.

On reviewing the current literature with respect to solid state crystal structures of

Ag(I)(NHC) complexes, it is apparent that it is difficult to predict the exact structure

of any given complex. This is due to the ability of Ag(I) to form complex anions of

the formula [AgX2]’ (X = Halogen), to be able to coordinate to either one or two NHC

moieties and engage in Agl Agl interactions in the solid state13. This has led to a

number of different structural arrangements, such as 3.2 (Figure 3.2), 3.610 and 3.78

(Figure 3.4) amongst others.

tBu *Bu

w

/ = \■Nv .N

- - A gb z ^ V - ^ b z Bf\ . - - " Br

I A g'

__ X „Br N N

3.6 3.7

Figure 3.4: Examples of structural diversity seen within Ag NHC complexes

As an illustration of the unpredictability of Ag complexes the crystal structural

evidence of Danopoulos for complexes 3.8 and 3.9 (Figure 3.5) are examined. 3.9 has

the carbene moiety directly attached to the pyridine making it a more rigid ligand than

that of 3.8. Introducing a second ligand to 3.9 to form the Ag(carbene) 2 complex

would most likely increase unfavourable steric interligand interactions. However

Danopoulos also observed that when the Ag NHC preparations were carried out at

higher temperatures (refluxing 1 ,2 -dichloroethane), by-products were formed which

were concluded to be the isostoichiometric Ag(ligand)Br (for 3.8) and

[Ag(ligand)2][AgBr2] for 3.9. The reasons why the different structures were adopted

were therefore inconclusive4.


R=2.6-Pri2C6H3

3.8 3.9

Figure 3.5: Danopoulos’s structurally characterised Ag NHC complexes.

It is important to note that in the examples in the literature, with the exception of

complexes prepared in the absence of coordinating halide ions, the stoichiometry is

mostly always 1:1:1 (NHC:Ag:X). Complexes prepared without the presence of

coordinating halide ions show a stoichiometry of 1:2 (Ag:NHC).

3.2 Results And Discussion.

3.2.1 Ag(l) Complexes Of Quinoline Functionalised Carbenes.

All of the quinoline functionalised imidazolium salts described in Chapter two were

successfully reacted with Ag2 0 to form the corresponding Ag(I) complexes. In a

typical reaction, two equivalents of the salt was taken up in DCM and stirred with one

equivalent of Ag2 0 for several hours at ~40°C, until the black suspension of Ag2 0

was replaced with the beige/grey coloured silver halide precipitate. Molecular sieves

were added to the reaction mixture as water is formed as a by product. The addition of

molecular sieves has also been shown to facilitate the formation of the Ag carbene

complex4. As silver complexes are light sensitive with respect to decomposition, the

reaction mixture was also protected from light. After the reaction was complete the


solutions were filtered to remove the silver halide, the solvents removed under

reduced pressure, and the beige Ag product washed with Et2 0 . The residues were1 1 1recrystallised from DCM and hexane and characterised by H and C NMR. H NMR

data clearly shows an absence of the characteristic C2 proton peak with no residual• • 1 signal remaining, which is a good indication that the reaction has proceeded. The C

NMR also shows the C-Ag peak within the 180ppm region. All of the complexes

showed good stability to air and moisture, though decomposition occurred in solution

over a period of time when exposed to light, again a good indication of Ag content.

Crystals suitable for X-ray diffraction were attempted by using a layering method of

DCM and hexane, though no suitable crystals could be grown.

Using the data collected thus far it is not possible to give the exact structure of the

silver complexes of these ligands. For illustrative purposes only, the structures of the

quinoline based Ag complexes are represented in a similar manner to the

octahydroacridine based complex which was crystalographically analysed, i.e one of a

mono NHC Ag complex with a coordinated anion.

The following complexes (3.10-3.13) were therefore synthesised and characterized by

lH and 13C NMR (Figure 3.6).

(T ) — Ag— Br f y — AgN

) Ag y — Ag—Br

3.10 3.11 3.12 3.13

Figure 3.6: Ag complexes of the quinoline functionalised NHCs.


3.2.2 Ag(I) Complex Of Octahydroacridine Functionalised

Imidazolium Salt

The Ag2 0 route was successfully applied to the octahydroacridine functionalised

imidazolium salt to give the Ag(I) complex. Again the reaction was carried out in a

similar manner to the quinoline functionalised salts, though a higher temperature was

used (~45°C) overnight. The tan coloured Ag complex proved to have similar

stabilities to the quinoline functionalised ones, with slow decomposition in solution.

lH NMR revealed an absence of the imidazolium C2-H proton at 10.15ppm, with no

residual signal remaining, thus concluding that complex 3.14 (Figure 3.7) was

successfully synthesised. Crystals suitable for X-ray diffraction were grown by adding

hexane onto a DCM solution of the complex (protected from light). The structure and

numbering scheme of 3.14 is given in Figure 3.8. Selected bond lengths and angles

are given in Table 3.1.

— Ag—Cl

3.14

Figure 3.7: Ag complex of the octahydroacridine functionalised NHC.


C 1 3

H q C14^ \ C 1 5

\ _ ^ O C 1 6

C 1 2C 25

C 24 C 19

C 1 8C 11

C 2 0

C 3 C1 C21C 4C 1 0 1C 2 2

C 1 7 N2C 9 C 2 3C 2

C 5C 8

Ag1C 6

C 7CM

X

Figure 3.8: ORTEP representations of 3.14 (50% probability of thermal

ellipsoids). Hydrogen atoms omitted for clarity.


Cl N1 1.446(4) C4 N2 1.382(4) N2 C2 N1 104.0(2)

C2 N2 1.344(4) C5 N2 1.471(4) N2 C2 Agl 126.3(2)

C2 N1 1.355(4) C5 C17 1.526(4) N1 C2 Agl 129.3(2)

C2 Agl 2.084(3) C17 N3 1.345(4) C2 N1 Cl 124.8(3)

C3 C4 1.343(5) Agl Cll 2.3395(8) C2 N2 C5 123.8(2)

C3 N1 1.384(4) C2 Agl Cll 177.21(8)

Table 3.1: Selected bond lengths (A) and angles (°) for 3.14.

Complex 3.14 is shown to be a monomer with a silver atom coordinated by a carbene

carbon and a chloride in a quasi-linear geometry. The C2-Ag-Cl bond angle of

177.21(8)° is comparable to 3.9 (176.1(2)°). The Ag-C2 bond length (2.084(3)A) is

also similar to that of 3.9 (2.075(7)). Like the parent imidazolium salt, the plane of the

NHC ring adopts a position perpendicular to the planes of the octahydroacridine and

mesityl moieties thereby minimising steric interactions. The dihedral angle between

the plane of the NHC and the octahydroacridine is approximately 107.1°, this is in

contrast to the angle of 31.5(3)° formed by the imidazole against the pyridine of 3.9

and reflects the greater steric demands of the octahydroacridine. The N1-C2-N2 bond

length and angles are all within the range of other NHC Ag complexes2’3,4.

3.2.3 Ag(I) Complex of the Bis-imidazolium Salt DIOP Analogue

(2.27)

Initial attempts to synthesis the Ag complex of the bis-carbene were unsuccessful and

the unreacted salt was recovered. Exchanging the hexafluorophosphate counter ion for

iodide by the addition of two equivalents of KI to the reaction mixture however,

promoted the reaction to proceed. One equivalent of the salt was reacted with one

equivalent of Ag2 0 in acetonitrile at 40°C overnight and the resulting beige

precipitate was filtered off and the solvent removed under reduced pressure to leave

an off white solid. The product was then washed with Et2 0 and recrystallised from1 12acetonitrile. The complex was characterized by H and C NMR and crystals were

grown by cooling an acetonitrile solution of the complex though a crystal structure


was not obtained. The main evidence for Ag carbene formation is the absence of the

characteristic imidazolium C2 proton and the appearance of a peak at 179.2ppm in the13 1C spectra corresponding to C-Ag. The H NMRs for the imidazolium salt and

corresponding Ag complex are given in Figure 3.10. The complex showed slow

decomposition in solution when exposed to light, again further evidence of Ag content.

For illustrative purposes and in the absence of a crystal structure, the complex is

shown to be a dinuclear Ag bridging complex such as 3.15 (Figure 3.9). This is based

around other examples9,14"17 where linkers connecting the NHC functions are

sufficiently long and flexible to allow for this arrangement.

Agl

3.15

Figure 3.9: Proposed dinuclear Ag bridging NHC structure.


F (6 H)

D (4 H)

A (2H) B (4 H) C (2 H)

s e

Agl,

C (6 H)

D (6 H)

A (4 H)

i i

Figure 3.10: NMR spectra for the DIOP analogue imidazolium salt and its Ag(I)

complex.


3.3 Conclusions

In conclusion, the complete range of imidazolium salts synthesised in chapter two

were successfully reacted with Ag2 0 to form the corresponding Ag(I) complexes.

This conclusion was drawn from the NMR data which showed that the imidazolium

C2-H proton was absent, usually a sufficient indication that a complex was

synthesised. A crystal structure of the octahydroacridine functionalised complex was

also obtained as evidence of this, and showed that the structure was one of an NHC-

Ag-X arrangement. All complexes showed stability to air and moisture though slow

decomposition in solution did occur when exposed to light.

3.4 Experimental

3.4.1 General Comments




(THF), diethyl ether (Et2<3 ), and hexane were dried and degassed by refluxing under

dinitrogen over sodium wire and benzophenone. Dichloromethane (DCM), methanol

(MeOH), and acetonitrile (MeCN) were dried over calcium hydride. All other

anhydrous solvents were obtained by distillation from the appropriate drying agents

under dinitrogen. Deoxygenation of solvents and reagents was carried out by freeze

thaw degassing.


sieves and freeze-thaw degassed three times. All reagents were purchased from


Chapter Three: Ag(I) Complexes Of Chelating Carbenes 87

All NMR data are quoted in 5/ppm. *H and 13C spectra were recorded on a Bruker 400

MHz DPX Avance, unless otherwise stated, and referenced to SiMe4 . Elemental

analysis was carried out by Warwick Analytical Service Ltd, Coventry (UK).

Synthesis of [Ag(l-methyl-3-quinoline-imidazolin-2-ylidene).I] (3.10):

N A g-I

N

Ag2 0 (0.152g, 0 .6 6 mmol) was added to a mixture of 1 -methyl-3-quinoline-

imidazolium iodide (0.444g 1.31 mmol) and 4A molecular sieves in 1 0 ml DCM to

give a black suspension. The mixture was protected from light and stirred overnight at

40°C. The resulting Agl was removed by filtration through celite and the solvent

removed under reduced pressure to leave a tan coloured solid which was washed with

Et2 0 (2x10ml). The solid was then recrystallised from DCM/Hexane to give the

desired Ag complex 3.10 (0.26g, 63%). lH NMR (d2-DCM, 250MHz, 5) 8.80 (m, 2H,

Quin H2), 8.20 (m, 2H, Quin H4), 7.95 (m, 2H, Quin H6 ), 7.80 (m, 2H, Quin H8 ),

7.4-7.5 (m, 4H, QuinH3,7), 7.35 (m, 2H, HCCH), 7.15 (m, 2H, HCCH), 3.85 (s, 6 H,

NCH3) 13C NMR (d2-DCM, 100MHz, 5): 182.0 (C-Ag) , 163.2, 151.3, 147.6, 145.2,

139.4, 138.5, 138.1, 136.9, 134.7, 133.2, 131.8, 44.0 (CH3).

Synthesis of [Ag(l-benzyl-3-quinoline-imidazolin-2-ylidene)Br] (3.11):


To a mixture of 1-benzyl-3-quinolinimidazolium bromide (2.69g, 0.787mmol) and 4A

molecular sieves in 10ml DCM, Ag2 0 (0.091 g, 0.394mmol) was added to give a black

suspension which was protected from light and stirred overnight at 40°C. The

resulting beige precipitate was removed by filtration through celite and the solvent

removed under reduced pressure to leave a brown coloured solid. The solid was

washed with Et2 0 (3x10ml) and recrystallised from DCM/Hexane to give the desired

Ag complex (3.11). (0.132g, 47%). *H NMR (CDC13, 250MHz, 8): 8.85 (m, 2 H,

Quin H2), 8.25 (m, 2H, Ar H), 7.90 (m, 4H, ArH), 7.55 (m, 2H, ArH), 7.45 (m, 4H,

HCCH), 7.25-7.35 (m, 10H, ArH), 7.10 (2H, m, ArH), 5.35 (s, 4H, CH2-Ph). 13C

NMR (d2-DCM, 100MHz, 8): 183 (NCAg) 151.7, 142.9, 137.1, 136.9, 136.3, 130.0,

129.8, 129.7, 129.5, 129.2, 128.8, 128.4, 126.5, 125.9, 122.8, 120.9, 56.1(CH2)

Synthesis of [Ag(l-methyl-3-(2-methyl-quinoline)-imidazoIin-2-ylidene)I] (3.12):

Ny Ag-I

N

1-methyl-3-(2-methyl-quinoline)-imidazolium iodide (0.3g, 0.85mmol) was dissolved

in 10ml DCM with 4A molecular sieves. Ag20 (O.lg, 0.43mmol) was added to the

mixture to give a black suspension that was protected from light and stirred at 40°C

overnight. The resulting beige precipitate was removed by filtration and the solvent

removed under reduced pressure to leave a brown coloured solid. The solid was

washed with Et20 (2x10ml) and recrystallised from DCM/Hexane to give the desired

Ag complex (3.12) (0.12, 43%). lU NMR(CDC13, 500MHz, 8 ): 8.10 (m, 2H, Quin H),

7.65-7.85 (m, 4H, Quin H), 7.35-7.45 (m, 4H, Quin H), 7.15-7.25 (m, 4H, HCCH),

3.75 (s, 3H, NCH3), 2.55 (s, 3H, Quin CH3). 13C NMR (CDC13, 100MHz, 8 ): 184.0

(NCN), 161.0, 142.0, 137.5, 137.0, 129.0, 128.0, 127.0, 125.0, 124.8, 123.5, 122.0,

39.5 (NCH3), 25.5 (Quin CH3).

Synthesis of [Ag(l-benzyl-3-(2-methyl-quinoline)-imidazolin-2-ylidene)Br] (3.13):


Ag2 0 (0.095g, 0.41 mmol) was suspended in a solution of 1-benzyl-3-(2-methyl-

quinoline)-imidazolium bromide (0.350g, 0.82mmol) in 10ml DCM with 4A

molecular sieves. The black suspension was protected from light and stirred overnight

at 40°C. The resulting precipitate was removed by filtration and the solvent removed

under reduced pressure to leave a brown coloured solid. The solid was washed with

Et2 0 and recrystalised from DCM/Hexane to give the desired Ag complex 3.13.

(0.19g, 56%). *H NMR (CDC13, 250MHz, 8 ): 8.10(m, 2H, ArH), 7.85 (m, 4H, ArH),

7.55 (m, 4H, ArH), 7.45 (m, 2H, ArH), 7.35 (m, 10H, ArH), 7.10(m, 2H, ArH), 5.4

(4H, CH2-Ph), 2.65 (s, 6 H, Quin CH3). 13C NMR (DCM, 100MHz, 8): 178.5 (NCN),

159.7, 141.2, 137.1, 135.6, 135.2, 135.1, 128.3, 128.2, 127.8, 127.2, 127.1, 126.7,

125.8, 124.8, 124.2, 122.4, 119.3, 55.0 (CH2-Ph), 24.7 (Quin CH3).

Synthesis of [Ag(l-mesityl-3-octahydroacridinimidazolin)Cl] (3.14):

Ag2<3 (0.255g, 0.97mmol) was suspended in a solution of a l-mesityl-3-

octahydroacridinimidazolium chloride (0.79g, 1.94mmol) and 4A molecular sieves in

10ml DCM. The reaction mixture was protected from the light and stirred at 45 °C

overnight. The resulting AgCl was filtered off and the solvent removed under reduced

pressure to leave a yellow-brown powder. This was repeatedly washed with Et2 0

(3x10ml) and recrystallized from DCM/Hexane to give the silver complex (3.14).

Chapter Three: Ag(I) Complexes Of Chelating Carbenes 90

(3.8g, 44%). Crystals suitable for X-ray diffraction were grown by layering DCM and

Hexane. 'H NMR (CDCI3 ,400MHz, 8 ) 7.1 (m, 1H, ArH), 6.9 (m, 3H, ArH), 6.85 (m,

1H, ArH), 1.5-2.95 (m, 24H, CH2, CH3). ,3C NMR (CDCI3,100MHz, 8):185.9(NCN),

148.9, 139.2, 137.9, 135.9,135.0, 134.7, 132.4,130.3, 129.3, 121.6,120.7,61.8,40.2,

32.6,32.1,28.5,27.9,23.1,22.6,21.1,19.9,17.7.

Synthesis of chiral DIOP analogue Ag complex (3.15):

Agl.

4,5-Bis(methylimidazolium hexaflurophosphate)-2,2-dimethyl-1,3-dioxolan (0.246g,

0.42mmol) was dissolved in CH3CN with 4A molecular sieves. Ag2 0 (0.097g,

0.42mmol) and KI (0.069g, 0.42mmol) was added to the mixture giving a black

suspension which was protected from light and stirred at 40 °C overnight. The solution

was then filtered through celite to remove precipitated AgX and the solvent removed

under reduced pressure to give a solid. This was washed with Et2 0 to give 3.15 as an

off white solid. (0.21g, 39%). *H NMR(d6-DMSO, 400MHz, 8 ): 7.5 (m, 8 H, HCCH),

4.6 (m, 4H, OCH), 4.45 (m, 4H, CH2), 4.35 (m, 4H, CH2), 3.85 (s, 12H, NCH3), 1.35

(s, 12H, CCH3). 13C NMR (de-DMSO 100MHz, 8 ): 179.2 (NCN), 122.0 (NCCN),

121.6 (NCCN), 109.5 (OCC2H6), 76.8 (OCH), 51.8 (CH2), 37.1 (NCH3) 26.10 (CH3).

Elemental Anal. Calc. C30H44N4O4Ag2l2 (1280.9): C 28.1, H 3.4, N 8.7. Found: C

29.2, H 3.6, N 8.3.


3.5 References1. D. McGuinness, K. Cavell, Organometallics, 2000,19, 741-748.

2. A. Arduengo, R. Dias, J. Calabrese, F. Davidson, Organometallics, 1993,12,

3405-3409.

3. H. Wang, I. Lim, Organometallics, 1998,17, 972-975.

4. A. Tulloch, A. Danopoulos, S. Winston, S. Kleinhenz, G. Eastham, J. Chem.

Soc., Dalton trans., 2000,24, 4499-4506.

5. A. Magill, D. McGuinness, K. Cavell, G. Britovsek, V. Gibson, A. White, D.

Williams, A. H. White, B Skelton., J. Organomet. Chem., 2001 617-618, 546-

560.

6 . B. Bildstein, M. Malaun, H. Kopacka, K. Wurst, M. Mitterbock, K. Ongania,

G. Opromolla, P. Zanello, Organometallics, 1999,18,4325-4336.

7. C. Lee, J. Chen., K. Lee, C. Liu, I. Lin, Chem. Mater., 1999,11(5), 1237-1242.

8 . C. Lee, K. Lee, I. Lin, Organometallics, 2002,21, 10-12.

9. D. Nielsen, K. Cavell, B. Skelton, A. H. White, Inorg. Chim. Acta, 2002, 327,

116-125

10. J. Pytkowicz, S. Roland, P. Mangeney, J. Organomet. Chem., 2001,631, 157-

163.

11. W. Chen, B. Wu, K. Matsumoto, J. Organomet. Chem., 2002, 654 233-236.

12. L. Bonnet, R. Douthwaite, R. Hodgson, Organometallics (comms), 2003,

22(22), 4384-4386.

13. D. Nielsen, Ph.D Thesis.

14 D. Nielsen, K. Cavell, B. Skelton, A. White, Inorg. Chim. Acta., 2003,352,

143-150.

15 J. Garrison, R. Simons, C. Tessier, W. Youngs, J. Organomet. Chem., 2003,

673, 1-4.

16 J. Wang, H. Song, Q. Li, F. Xu, Z. Zhang, Inorg. Chim. Acta. 2005,358,

3653-3658.

17 A. Melaigue, Z. Sun, K. Hindi, A. Milsted, D. Ely, D. Reneker, C. Tessier, W.

Youngs, J. Am. Chem. Soc., 2005,127, 2285-2291

Chapter Four: Palladium Carbene Complexes.

Chapter Four

Palladium Carbene Complexes

4.1 Introduction- NHC metal complexes.

A multitude of mixed donor NHC metal complexes have been synthesised by various

groups, tethering a range of secondary donors to the NHC including amines, alkoxides

and phosphines amongst others. These have all been synthesised with similar aims to

those of this work, such as the ability to vary the steric bulk, chelate rigidity and

electronic asymmetry. A number of examples of these mixed donor complexes are

given in Figure 4.1.

/ = \ Me s — ^

Cl— P d -N ^ICl

4.11

'Pr\

/ = \ - N L . N

/ /— Pd-N

Ar

4.22

O.M es'

N

r= \-N n ^ n

•cu - Y LN—f O \ Ph-— --O —C u-0U .X TX-

WMes

•Pr

4.55

Ph

Pd Ph•Pr

4 .33

Ph

Mes

4 .66

to ^‘Bu—

•Bu \ } Br <Bu

rNlN N— ‘Bu\= J

4.44

\ /^—N Ph ^ P d \ Br Br Ar

4.88

Figure 4.1: Examples of functionalised NHC metal complexes.

The first Pd(II) carbene complexs were reported by Herrmann in 19959 by the reaction

of Pd(OAc) 2 with 1,3-dimethylimidazolium iodide to give 4.9 and 3,3-dimethyl-1 ,1 -

methylenediimidazloium diiodide to give 4.10 (Scheme 4.1). Herrmann showed that

these new Pd complexes were ‘extraordinarily stable to heat, oxygen and moisture’.


2 [ © ) + Pd(°Ac)2 N

N /-2 AcOH

f Y■n^ p i 4.9

< + Pd(OAc) 2 4.10

\

-2 AcOH

\

Scheme 4.1: Herrmann’s synthesis of 4.9 and 4.10

Further to this, Herrmann also demonstrated these Pd carbene complexes to be

excellent catalysts in the Heck reaction, and in turn offered a viable alternative to the

ubiquitous phosphines.

Depending on the nature of the ligand precursor there are a number of routes to

synthesise Pd(NHC) complexes and these have been outlined in Chapter one along

with other metals. The classical route is by reaction of an imidazolium salt with

Pd(OAc)2 (Scheme 4.1). For many simple mono and bis imidazolium salts this

procedure is usually satisfactory and produces the Pd complex in moderate to high

yields. However it suffers draw backs in that this route doesn’t allow access to the

catalytically interesting Pd-hydrocarbyl species and the solvent of choice is usually

DMSO which can prove to be problematic. The high temperatures often associated

with this route can also be disadvantageous in some systems where thermal stability is

an issue. Another common route is via the free carbene which can be generated from

imidazolium salts via a number of different bases including NaH with catalytic

amounts of KOlBu and amides such as K[N(SiMe3)2]. It can also be prepared from

imidazole-2-thiones via abstraction of the sulphur atom with potassium. Once formed,

the free carbene can be directly reacted with an appropriate metal source such as

PdCbCCOD). Cavell demonstrated this method to be of use in the formation of the


important Pd-hydrocarbyl species. Methylating Pd carbene complexes with various

alkylating agents had previously proved unsuccessful and so it was shown that

reaction of l,3-dimethylimidazolin-2-ylidene with PdMeCl(COD) led to the formation

of a methylpalladium carbene chloro-bridged dimer (4.11)10 (Scheme 4.2). This was

found to be a valuable precursor for the preparation of a range of other

methylpalladium complexes, both neutral and cationic. Cavell for example, reacted

4.11 with thallium 2-pyridinecarboxylate to give 4.12 in 100% yield10.

4.11 4.12

Scheme 4.2: Cavell’s route to Methylpalladium(II) complexes.

Where NHCs do not posses sufficiently bulky aryl or alkyl groups on the 1 and 3

positions of the ring, the free carbene route may be unsuitable as these provide

stability to the free carbene and thereby aid complexation. It has also been found that

the use of strong bases may cause problems if functionalised carbenes are used, due to

acidic protons commonly associated with the functional group and any CH2 linkers.

Oxidative addition of low valent metal precursors to a carbon-hydrogen bond has also

been shown11,12 and has successfully been applied with Pd2(dba) 3 and imidazolium

salts13.

As mentioned in Chapter three, the Ag2 0 route was seen as the most appropriate route

for functionalised carbene hydrocarbyl-Pd complex formation and was applied to the

salts synthesised in this work. Alternative routes were attempted and will be discussed

later.

4.2 Results and Discussion

c ' -CODC: + PdMeCI(COD) --------► 1/2

N\

4.2.1 Quinoline Functionalised Carbene Complexes Of Pd


In a typical reaction, the quinoline functionalised Ag(I)(NHC) was reacted with one

equivalent of PdMeCl(COD) in DCM for several hours, the PdMeCl(COD) being

synthesised by standard procedures. The Pd complexes were characterised by lH and

13C NMR. The characteristic peak in the NMR is the appearance of a singlet at

~0.55-0.75ppm corresponding to the Pd-Methyl protons. The appearance of the

imidazolium C2 protons was not observed. The following Pd(II)NHC complexes

based around a quinoline system were synthesised (4.13-4.16, Figure 4.2). Crystals

suitable for X ray diffraction were grown for complex 4.13 by layering hexanes on to

a DCM solution of the complex. The structure and labelling scheme of complex 4.13

is given in Figure 4.3 with selected bond lengths and angles in Table 4.1. All of the

complexes synthesised showed good stability to air and moisture.

N

-Ny A g-l

N

PdMeCI(COD)

DCM 40C

N

Pd-CII

Me

COD

Agl

3.10 4.13

Pd-CII

Me

N'

Pd-CII

Me

4.14 4.15

Pd-CIMe

4.16

Figure 4.2: Palladium complexes of the quinoline functionalised carbene’s

Complex 4.13 exhibits a square planar geometry that is slightly distorted where the

chelating ligand occupies the two cis positions. The six membered ring formed by the

chelate is slightly twisted with a dihedral angle between the plane of the quinoline and


the plane of the imidazolium ring of approximately 35.5°. This is consistent with the

observation made in Chapter two from the crystal structure of imidazolium salt 2.8.

Similar six membered chelates such as Danopoulos’s (4.17, Figure 4.4) 14 have been

shown to form a puckered confirmation in order to minimise conformational strain,

which is achievable due to the sp3 hybridized CH2 linker. The rigidity of 4.13 imposed

by the aromatic ring system minimises deviations from overall ligand planarity and is

therefore more comparable to five membered chelates such as 4.1814 where the

chelate is relatively planar. The inclusion of the methyl group in the C12 position of

complexes 4.15 and 4.16 should cause the chelate to be more strained. The carbene

moiety of 4.13 is trans to the chloride which is consistent with other examples. The

bond lengths around the metal centre are not unusual for this type of complex14,16. The

ligand forms a bite angle of 86.9(2)°, slightly larger than that of the six membered

chelate (4.17, 85.22(13)A) and the five membered chelate (4.18, 79.15(3)A)

CM

Figure 4.3: ORTEP representation of the quinoline functionalised carbene complex

4.13, (50% probability of thermal ellipsoids). Hydrogen atoms are omitted for clarity.


Figure 4.3 cont.: ORTEP representation of the quinoline functionalised

carbene complex 4.13, (50% probability of thermal ellipsoids). Hydrogen atoms are

omitted for clarity.

C2 N1 1.366(8) C13 C5 1.410(8) C2 N1 Cl 125.9(5)

C2 N2 1.380(7) C13 N3 1.376(7) C2 N2 C5 125.8(5)

C4 C3 1.321(9) Pdl C2 1.943(6) C2 Pdl Cll 176.15(17)

C4 N2 1.412(8) Pdl N3 2.169(5) C14 Pdl N3 173.7(2)

C3 N1 1.392(8) Pdl C14 2.050(6) C2 Pdl C14 93.7(2)

Cl N1 1.443(8) Pdl Cll 2.3763(14) C2 Pdl N3 86.9(2)

C5 N2 1.427(8) N1 C2 N2 103.5(5) C14 Pdl Cll 88.88(17)

N3 Pdl Cll 90.28(13)

Table 4 .1 : Selected bond lengths (A) and angles (°) for complex 4.13


4.17 4.18

Figure 4.4: Danopoulos’s five and six membered chelating ligands14.

4.2.2 Pd Complexes Of An Octahydroacridine Functionalised Carbene.

The Ag2 0 route was again used for synthesising the Pd complexes of the

octahydroacridine functionalised carbene. In a similar manner to the quinoline

functionalised complexes, the Ag(I) complex of the octahydroacridine carbene was

reacted with one equivalent of either PdMeCl(COD) or PdC^CCOD) in DCM and

stirred for several hours at room temperature. The solution was filtered to remove the

precipitated AgCl and the solvent removed under reduced pressure. The yellow solids

were washed with Et2 0 to remove any residual COD and were recrystallised from

DCM and Hexane. Both the dichloro (4.19) and methyl chloro (4.20) Pd complexes1 1 *3were synthesised (Figure 4.5) and characterized by H and C NMR. Crystals suitable

for X-ray diffraction were grown by layering hexane onto a DCM solution of both

complexes. Both complexes showed good stability to air and moisture, the crystals of

4.20 being exposed to air for several months with little decomposition. The crystal

structures and numbering schemes for complexes 4.19 and 4.20 are given in Figure

4.6 and 4.7, with selected bond lengths in tables 4.2 and 4.3 respectively.

4.19 4.20

Figure 4.5: Pd complexes of the octahydroacridine functionalised NHC


Figure 4.6: ORTEP representations of complex 4.19 (50% probability of thermal

ellipsoids). Hydrogen atoms are omitted for clarity.


C2 N1 1.352(4) C17 N3 1.357(5) C2 Pdl C12 172.17(11)

C2 N2 1.347(5) C2 Pdl 1.949(4) N3 Pdl Cll 173.63(9)

C3 N1 1.395(5) N3 Pdl 2.060(3) C2 Pdl N3 83.40(13)

C4 N2 1.400(4) Cll Pdl 2.3036(9) C2 Pdl Cll 92.10(11)

C3 C4 1.334(5) C12 Pdl 2.3748(9) N3 Pdl C12 91.68(8)

Cl N1 1.445(5) N2 C2 N1 106.3(3) Cll Pdl C12 92.30(3)

C5 N2 ’ 1.488(5) C2 N1 Cl 127.3(3)

C5 C17 1.524(5) C2 N2 C5 120.4(3)

Table 4.2: Selected bond lengths (A) and angles (°) for complex 4.19.

C8 C7

C13

C 9 C6C10

C5C11 C17’C12 C4

N2'N3C16 Pd1 hC3,C2

N 1CI1C15C14 C26 C24C1

C18C23C22

C19C21 C20

C25

Figure 4.7: ORTEP representation of complex 4.20 (50% probability of thermal



Figure 4.7 cont.: ORTEP representation of complex 4.20 (50% probability of thermal


C2 - N1 1.354(4) C17- N3 1.355(4) C2 - Pdl - Cll 174.06(10)

C2 - N2 1.354(4) C2 - Pdl 1.975(3) C26 - Pdl - N3 174.96(12)

C3 - N1 1.403(4) N3 - Pdl 2.224(3) C2 - Pdl - C26 91.49(14)

C4 - N2 1.384(4) C26 - Pdl 2.030(4) C2 -Pdl -N3 84.73(13)

C3 - C4 1.325(5) Cll - Pdl 2.3814(8) C26 -Pdl -Cll 88.46(10)

Cl - N1 1.448(5) N2 - C2 - N1 104.80(3) N3 - Pdl - Cll 94.96(7)

C5 - N2 1.496(5) C2 - N1 - Cl 126.3(3)

C5 - C17 1.514(5) C2 - N2 - C5 119.8(3)

Table 4.3: Selected bond lengths (A) and angles (°) for complex 4.20.

Complexes 4.19 and 4.20 both exhibit square planar geometries which are slightly

distorted around the Pd Centres. The chelating ligands occupy the cis positions in both

cases. The unit cell contains the R and S isomers in both cases. Figure 4.6 illustrates

complex 4.19 containing the R enantiomer of the carbene ligand whilst figure 4.7

shows complex 4.20 containing the S. Both six membered chelate rings adopt a boat


configuration which is made possible by the sp3 hybridized linker and is comparable

to 4.17. 4.19 has a bite angle of 83.40(13)° and 4.20 has an angle of 84.73(13)°. Both

are smaller than the quinoline functionalised complex (4.13) and Danopoulos’s (4.17)

and are an indication of how different linkers can effect bite angle. The plane of the

imidazolium ring and the plane of the octahydroacridine rings form a dihedral angle

of approximately 60.7° in complex 4.19 and approximately 69.1° for complex 4.20.

The larger angle for 4.20 has an effect of reducing interligand steric interactions by

increasing the distance between the Pd-Methyl group (C26) and the methyl groups of

the mesityl moiety, the closest in the solid state structure being C24 with a distance of

3.748A. The distance between Cll and C23 in 4.19 is 3.646A. The N3-Pdl bond

lengths of complexes 4.19 and 4.20 are 2.060(3)A and 2.224(3)A respectively the

latter being longer due to the stronger trans influence of the methyl group on 4.20

compared to Cl. Again the Pd-Methyl group is cis to the NHC in complex 4.20. In

complex 4.19, the plane of the imidazole forms an angle of approximately 89.6° with

the plane of the mesityl moiety, and an angle of 49.8° with the plane of the four

coordinate Pd centre. The plane of the octahydroacridine function forms an angle of

57.1° with the plane of the four coordinate Pd centre. In complex 4.20 these angles are

approximately 83.3°, 53.1° and 50.7° respectively.

4.2.3 Electrospray Mass Ionisation (ESI) Spectroscopy of 4.14 & 4.20

High resolution electrospray mass spectrometry was carried out on 4.14 and 4.20 in

order to confirm their identity which also provided an insight into their reactivity. The

mass spectroscopy of these ligands was undertaken in the presence of an acetonitrile

mobile phase and in both cases chloride displacement by acetonitrile afforded cations

observable in the ESI +ve mode. Two types of ion products were observed, firstly

those in which acetonitrile simply displaces a coordinated chloride ligand and

secondly those which involved chloride displacement and subsequent metallation of

the carbene ligand framework. The later class of ion product, by necessity, requires

oxidative addition and reductive elimination steps and thus provides indirect evidence

for the potential for catalytic activity. The composition of the ligand precursors 2.7

and 2.14 were verified by the observation of the imidazolium parent cations and

requires no further discussion.


A different pattern of reactivity is observed for the Pd (II) complexes of the quinoline

and octahydroacridine based ligands. In the case of the hydrocarbyl complex 4.20

simple displacement of the chloride by acetonitrile is observed (Scheme 4.3). In

common with 4.20, the quinoline based [Pd(II)(NHC)(Me)Cl] 4.14 undergoes

chloride substitution to form a cationic Pd(II) complex [Pd(NHC)(Me)(MeCN)]+

which is observed as the major ion product. In contrast to 4.20, this

[Pd(II)(NHC)(Me)(MeCN)]+ further reacts to form the metallated one. Although it is

not possible to comment on the sequence of reactions that occur within the mass

spectrometer, a suggested sequence is given (Scheme 4.3).

Calculated for C28H35N4Pd (M - C1‘ + MeCN), 533.1897; observed, 533.1782 (100 %)

Calculated for C22H2 iN4Pd (M - C1‘ + MeCN), 447.0801; observed, 447.0771 (100 %). Calculated for C2 iH l7N4Pd, 431.0488 (M - Cl" + MeCN - CH4); observed, 431.0469 (39 %).

+MeCN

+

+MeCN

+ +

P dN = C -C H . o- metallationMethaneelimination

Pd-N=C-CH.

Scheme 4.3: Suggested reaction pathways for the Pd(II) based complexes.


4.2.4 Pd Complex Of A Carbene DIOP Analogue

In order to generate the free carbene, deprotonation of the bis-imidazolium salt (2.23)

was attempted with a number of bases. None of the products however could be

isolated and verified as the desired free carbene. Again the Ag2<D route was therefore

seen as the best option for generating the Pd complex. While there were no major

problems associated with generating the Ag(I) complex, transmetallation to form a

stable Pd complex was not as successful. The silver complex was reacted with a

solution of PdMeCl(COD) at room temperature for several hours. The resulting AgCl

was removed by filtration and the solvents removed under reduced pressure to give a

white solid. After a short period of time the solid began to darken with a rapid colour

change towards a black/dark grey solid on washing with Et2 0 . lH NMR gave

consistently poorly resolved spectra in a number of different deuterated solvents,

though the reappearance of the imidazolium C2-H was not observed. It is possible that

the Pd complex forms but undergoes reductive elimination to form imidazolium salts

and Pd black. Although chelating ligands have been shown to stabilise metal

complexes from reductive elimination by reducing the bite angle this particular 9

membered ring chelate may be inefficient at providing a bite angle sufficient for this.

1 7During the course of this work Harrison had also synthesised an NHC DIOP

analogue and had successfully formed the Pd(II) complex (4.21, Figure 4.8). 4.21 was

formed in a 61% yield by reaction of the dibromo imidazolium salt with Pd(OAc)2 in

DMSO using a temperature gradient, and was found by crystal structure determination

to be the cis bis-carbene. Further to this Harrison also synthesised a similar

imidazolium salt with two extra CH2 spacers, which upon coordination gave the

eleven membered chelate trans bis-carbene (4.22, Figure 4.8) by the same method

(also crystalographically determined). Reaction with Pd(OAc) 2 with our salt (2.23)

was attempted using similar reaction conditions as well as the addition of KI to

exchange the counter ion. *H NMR revealed only poorly resolved spectra of the

imidazolium salt. Due to time constraints and in light of Harrison’s work, the

investigations into the formation of the desired Pd-hydrocarbyl species and its

possible subsequent reductive elimination were abandoned.


o

o

N

N

N

Pd:

Cy

.Br

" B r

N -C y

N -C y

OBr— Pd-Br

O

4.234.21

Figure 4.8: Harrison’s NHC DIOP analogue complexes.

4.3 Conclusions

Several Pd(II) complexes have been successfully synthesised using the Ag(I)

complexes of Chapter Three as transmetallation agents, the majority of these being the

catalytically interesting PdMeCl(NHC) species. All complexes showed good stability

to air and moisture. All the complexes were characterized by various means including

\H and 13C NMR with three of the complexes being crystalographically determined.

The crystal structure of 4.13 indicates the quinoline based systems to be relatively

planar chelates, with differences in the bond lengths of the Pd substituents reflecting

the electronic asymmetry of the chelating ligand and the trans effect. Both of these

features were targets in the ligand design. From the crystal structures of 4.19 and 4.20

the octahydroacridine system extends the ring bridging NHC-functional group system

by introducing a greater extent of flexibility to the chelate as well as a chiral centre.

Again electronic asymmetry is maintained for this ligand fulfilling some of the

objectives of the ligand design.

The Pd complex of the chiral bis carbene was unfortunately not isolated, possibly due

to a reductive elimination pathway. Performing the transmetallation procedure at


lower temperatures and maintaining these temperatures for NMR studies would be the

first step in investigating any reductive elimination pathway.

4.4 Experimental

4.4.1 General Comments




(THF), diethyl ether (Et2 0 ), and hexane were dried and degassed by refluxing under

dinitrogen and over sodium wire and benzophenone. Dichloromethane (DCM),

methanol (MeOH), and acetonitrile (MeCN) were dried over calcium hydride. All

other anhydrous solvents were obtained by distillation from the appropriate drying

agents under dinitrogen. Deoxygenation of solvents and reagents was carried out by

freeze thaw degassing


sieves and freeze-thaw degassed three times. All reagents were purchased from


All NMR data are quoted in 8 /ppm. ]H and 13C spectra were recorded on a Bruker 400

MHz DPX Avance, unless otherwise stated, and reference to SiMe4 . Electrospray

mass spectrometry (ESMS) was performed on a VG Fisons Platform II instrument by

the Department of Chemistry, Cardiff University. High accuracy electrospray mass

spectra were recorded on a Waters Micromass Q-Tof Micro spectrometer and are

referenced against a raffinose internal ion (lockspray) source. A Waters Acquity

UHPLC with a MeCN: H2O (90:10) mobile phrase was used as an autosampler with

trace HCO2H (0.001%) being added to the mobile phase promote positive ion

formation. Elemental analysis was carried out by Warwick Analytical Service Ltd,

Coventry (UK).

Chapter Four: Palladium Carbene Complexes. 107

PdClMe(COD) and PdCl2(COD) were prepared by standard methods in bulk18. PdCb

was supplied by Johnson Matthey.

Synthesis of [Pd(Me)(l-methyl-3-quinoline-imidazolm-2-ylidene)Cl] (4.13):

I ) — Pd-CIL

A solution of [Ag(l-methyl-3-quinoline-imidazolin-2-ylidene)I] (0.350g, 1.12mmol)

and PdMeCl(COD) (0.316, 1.12mmol) was stirred together in DCM for 2 hours at

room temperature. The mixture was filtered through celite to remove precipitated

silver chloride and Pd black. The solvent was then reduced to ca. 10ml and hexanes

added to precipitate a pale yellow powder. The solvent was removed by the use of a

filterstick and the remaining solid was washed with Et2 0 (3x10ml). The powder was

recrystallized from DCM/Hexane to give the Pd complex (4.13) (0.18g, 44%).

Crystals suitable for X-ray diffraction were grown by layering hexane on DCM. 1H

NMR (CDC13, 400MHz, 8 ): 9.55 (m, 1H, Quin H2), 8.3 (m, 1H, Quin H), 7.8 (m, 2H,

Quin H), 7.6 (m, 1H, Quin H), 7.50 (m, 1H, Quin H), 7.3 (m, 1H, HCCH), 7.10 (m,

1H, HCCH), 3.90 (s, 2H, NCH3), 0.75 (PdCH3) 13C NMR (CDC13) 100MHz, 8 )

171.85 (NCN) 156.31, 139.0, 138.5, 135.0, 129.8, 128.8, 128.0, 126.6, 123.9, 122.3,

120.9, 38.5, -9.8. Elemental Anal. Calc, for Ci4Hi4N3ClPd: C 45.8, H 3.8, N 11.4;

found: C 44.6, H 3.8, N 10.35

Synthesis of [PD(Me)(l-benzyI-3-quinoline-imidazoIin-2-ylidene)Cl] (4.14):

Pd-CII

Me

Chapter Four: Palladium Carbene Complexes. 108

A solution of [Ag(l-benzyl-3 quinoline-imidazolin-2-ylidene)Br] (0.25g, 0.74mmol)

and PdMeCl(COD) (0.195g, 0.74mmol) was stirred together in DCM for 2 hours at

room temperature. The mixture was filtered through celite to remove precipitated

silver chloride and any Pd black. The solvent was then reduced to ca. 10ml and

hexanes added to precipitate a brown powder. This was filtered by use of a filter stick

and washed with Et2 0 (3x10ml). The powder was then recrystallized from

DCM/Hexane to give the Pd complex (4.14). (0.14g, 43%). 1H NMR (CDC13,

500MHz, 5): 9.5 (m. 1H, Quin H2), 8.25 (m, 1H, Quin H), 7.75 (m, 2H, Quin H),7.60

(m, 2H, HCCH), 7.55 (m, ‘H, Quin H), 7.30 (m, 5H, ArH), 6.95 (m, 1H, Quin H),

5.35 (s, 2H, CH2-Ph), 0.75 (s, 3H, PdCH3). ,3C NMR (DCM, 100MHz, 8): 172.7

(NCN), 156.7, 139.4, 139.0, 136.71, 135.2, 130.1, 129.6, 129.4, 128.7, 128.5, 128.3,

127.2, 124.1, 123.1, 123.0, 122.4, 122.1, 54.9, -8.5. MS (ESI) m/z (%) Calculated for

C22H2 iN4Pd (M - Cl' + MeCN), 447.0801; observed, 447.0771 (100 %)

Synthesis of [Pd(Me)( 1 -methyl-3-(2-methyl-quinoline)-imidazolin-2-ylidene)Cl]

(4.15):

Ny — Pd-CI

^ Me

A solution of [Ag(l-methyl-3-(2-methyl-quinoline)-imidazolin-2-ylidene)I] (O.lg,

0.3mmol) and PdMeCl(COD) (0.079g, 0.3mmol) was stirred together in DCM for 2

hours at room temperature. The mixture was filtered through celite to remove the

precipitated silver chloride and any Pd black. The solvent was then reduced to ca.

10ml and hexanes added to precipitate a brown powder. The solvent was removed

with a filter stick and the solid washed with Et2 0 (3x10ml). The solid was then

recrystallized from DCM/Hexane to give the Pd complex (4.15). (0.12g, 8 6 %) 1H

NMR (CDC13, 500MHz, 8 ): 8.1 (m, 1H, Quin 4), 7.75 (m, 1H, Quin 6 ), 7.65 (m, 1H

quin 8 ), 7.50 (m, 1H, Quin 7), 7.25 (m, 1H, Quin3), 7.2 (m, 1H, HCCH), 7.1 (m, 1H,

HCCH), 3.85 (s, 3H, NCH3), 3.0 (s, 3H, Quin CH3), 0.65 (s, 3H, PdMe). 13C NMR

(CDC13, 100MHz, 5): 179.0, 162.0, 144.2, 139.0, 138.8, 128.7, 128.4, 126.7, 124.6,

124.4,123.3, 121.5,38.6, 29.6, -9.8.


Synthesis of [Pd(Me)(l-benzyI-3-(2-methyl-quinoline)-imidazolin-2-ylidene)Cl] (4.16):

Pd-CIMe

A solution of [Ag(l-benzyl-3-(2-methyl-quinoline)-imidazolin-2-ylidene)Br] (0.2g,

0.48mmol) and PdMeCl(COD) (0.127g, 0.48mmol) was stirred together in DCM for 2

hours at room temperature. The mixture was then filtered through celite to remove the

precipitated silver chloride and Pd black. The solvent was reduced to ca. 10ml and

hexanes added to precipitate a brown powder. The solvents were removed by a filter

stick and the remaining powder washed with Et2 0 (3x10ml). This was then

recrystalized from DCM/Hexane to give the Pd complex (4.16). (0.18g, 76%). 1H

NMR (DCM, 500MHz, 5): 8.15 (m, 1H, ArH), 7.75 (m, 2H, ArH), 7.55 (m, 1H, ArH),

7.45 (m, 2H, ArH), 7.35 (m, 5H, ArH), 7.0 (m, 1H, ArH), 5.25 (s, CH2-Ph) 2.95 (s,

3H, Quin CH3), 0.55 (s, 3H, PdCH3). 13C NMR (DCM, 100MHz, 8): 177.5 (NCN),

152.5, 139.8, 137.6, 135.9, 135.7, 134.4, 128.1, 127.8, 127.6, 127.3, 127.2, 127.0,

125.1, 123.0, 122.8, 122.4, 121.1, 120.6, 53.42 (quinCH3), -9.8 (PdMe).

Synthesis of [Pd(l-mesityl-3-octahydroacridinimidazolin-2-ylidene)Cl2] (4.19):

Pd-CI

A solution of PdCl2(COD) (0.123g, 0.434mmol) in 10ml DCM was added to a

solution of [Ag(l-mesityl-3-octahydroacridinimidazolin)Cl] (0.4g, 0.434mmol) in


10ml DCM and stirred at room temperature for 2 hours. The solution was then filtered

through celite to remove the precipitated silver chloride and the solvent removed until

ca. 5ml remained. Hexane was then added, precipitating a yellow solid which was

filtered with a filter stick. The solid was recrystalized from DCM/Hexane and washed

with hexane to give a yellow powder of 4.19. (1.6g, 67%). Crystals suitable for X-ray

diffraction were grown by layering DCM and hexane. 1H NMR (CDCI3 , 400MHz, 6 )

7.2 (m, 2H, ArH), 6.9 (m, 2H, ArH), 6.75 (m, 1H, ArH), 4.2 (m, 1H, CHN), 1.5-2.90

(m, 23H, CH2, CH3). Elemental Anal. Calc, for C25H29N3Cl2Pd (548.85): C 53.6, H

5.3, N 7.5; found: C 51.4, H 5.3, N 6.3.

Synthesis of [Pd(Me)(l -mesityl-3-octahy droacridinimidazolin-2-ylidene)Cl]

(4.20):

Pd-CIIMe

A solution of [Ag(l-mesityl-3-octahydroacridinimidazolin)Cl] (0.432g, 0.77mmol)

and PdMeCl(COD) (0.203, 0.77mmol) was stirred together in DCM for 2 hours at

room temperature. The mixture was filtered through celite to remove the precipitated

silver chloride and any Pd black. The solvent was then reduced to ca. 10ml and

hexanes added to precipitate a pale yellow powder. The solvents were removed with a

filter stick and the solid washed with Et20 (3x10ml). The powder was recrystalized

from DCM/Hexane to give the Pd complex 4.20. (0.27g, 6 6 %). Crystals suitable for

X-ray diffraction were grown by layering Hexanes onto a DCM solution of the

complex. *H NMR (DCM, 500MHz, 8 ) 7.25 (m, 2H, HCCH), 7.05 (m, 2H, ArH),

6 . 8 (m, 1H, ArH), 4.1 (m, 1H, CHN), 1.55-2.95 (m, 23H, CH2, CH3), 0.0 (s, 3H,

PdCH3). ,3C NMR (DCM, 100MHz, 8):159.28 (NCN), 139.4, 139.2, 135.65, 135.6,

135.2, 134.4, 131.2, 129.5, 129.0, 122.3, 116.9, 59.5, 34.4,29.2,28.8,27.5,23.0,22.6,


21.2, 20.4, 18.6, 18.2, -10.1 Elemental Anal. Calc, for C26H32N3ClPd (528.43): C 59.0,

H 6.0, N 6.9; found: C 52.6, H 5.56, N 6.74. MS (ESI) m/z (%):Calculated for

C28H35N4Pd (M - CF + MeCN), 533.1897; observed, 533.1782 (100 %)

4.5 References:

1. K. Coleman, H. Chamberlayne, S. Tuberville, M. Green, A. Cowley, J. Chem.

Soc., Dalton Trans., 2003, 2917-2922.

2. M. Froseth, A. Dhindsa, H. Roise, M. Tilset, J. Chem. Soc., Dalton Trans.,

2003,4516-4524.

3 L. Bonnet, R. Douthwaite, Organometallies, 2003,22, 4187-4189.

4. P. Arnold, S. Mungur, A. Blake, C. Wilson, Angew. Chem. Int. Ed., 2003, 42

5981-5984.


Comm., 2004, 1612-1613.

6. C. Legault, C. Kendall, A. Charette, Chem. Comm., 2005, 3826-3828.

7. A. Waltman, R. Grubbs, Organometallics, 2004, 23, 3105-3107.

8. N. Tsoureas, A. Danopoulos, A. Tulloch, M. Light, Organometallics, 2003, 22,

4750-4758.

9. W. A. Herrmann, M. Elison, J. Fischer, C. Kocher, G. R. Artus, Angew. Chem.

Int. Ed. Engl.,. 1995, 34(21), 2371-2374.

10. M. Green, K. Cavell, B. Skelton, A. White, J. Organomet. Chem., 1998, 554 ,

175-179.

11. P. Fraser, W. Roper, F. Stone, J. Chem. Soc. Chem. Comm., 1974, 102.

12. D. McGuinness, K. Cavell, B. Yates, B. Skelton, A. White, J. Am. Chem. Soc.,

2001,123, 8317.

13. S. Grundemann, M. Albrecht, A. Kovacevic, J. Faller, R. Crabtree, J. Chem.

Soc. Dalton trans., 2002, 2163-2167.

14. A. Tulloch, S. Winston, A. Danopoulos, G. Eastman, M. Hursthouse. J. Chem.

Soc., Dalton Trans., 2003, 699-708.


15 F. Cotton, G. Wilkinson, Advanced Organic Chemistry, 5th Ed., Wiley

Interscience.

16 A. Tulloch, A. Danopoulos, R. Tooze, S. CafFerkey, S. Kleinhenz, M.

Hursthouse, Chem. Comm., 2000, 1247-1248.

17. C. Marshall, M. Ward, W. Harrison, Tetrahedron Lett., 2004, 45, 5703-5706.

18. R. Rulke, J. Emsting, A. Spek, C. Elsevier, P. Van Leeuwen, K. Vrieze, Inorg.

Chem., 1993, 32, 5769.

Chapter Five: The Heck Reaction

Chapter Five

Pd catalysed C-C Coupling reactions : The Heck

Reaction

5.1 Introduction

Due to time constraints, catalyst testing of the complexes synthesised in Chapter Four

was confined to the Heck reaction. The Heck reaction was chosen due to its

widespread use as ‘one of the basic tools of organic preparations’, and ‘a natural way

to assemble molecules’1. This chapter will focus on the Heck reaction with systems

involving pyridine-carbene type ligands. Aryl bromides and chlorides were chosen to

evaluate the necessary parameters to give high TONs (turn over number) and thus be

viable catalysts, particularly for the aryl chlorides.

5.2 Results and Discussion

Catalytic testing of the Pd(II) complexes, described in Chapter Four, has focused on

the Heck coupling of the activated aryl bromide 4-Bromoacetophenone with n-butyl

acrylate (Scheme 5.1) which was likely to give positive results for comparison with

other similar systems. A small selection of the complexes were also tested for their

activity for the coupling of 4-chlorobenzaldehyde with n-butyl acrylate. Sodium

acetate was chosen as the base due to its proven suitability in the Heck reaction with

regard to a wide variety of substrates ' . Tetrabutylammonium bromide was used as an

additive, as it has previously been shown that ammonium salts have a positive effect

on the Heck reaction9,10. All reactions were carried out under standardised conditions

at 120°C for 16 hours. The results of the catalyst testing can be seen in Table 5.1,

while Table 5.2 provides examples of other NHC complexes of various groups for

comparison in the Heck reaction.


M T / / B r +

Scheme 5.1: Heck coupling of 4-bromoacetophenone with n-butyl acrylate.

(Pd) O

Entry Pre catalyst LoadingMol% Aryl halide Coupling

partnerTime(h)

Conversion(%)(GC) TON

1

CCpOC5

-if0.5 4-

bromoacetophenonen-butylacrylate 16 32 64

2ccp

1 /N~n C Pd-< Jci 7 0.1 4-bromoacetophenone

n-butylacrylate 16 50 500

3

OCp 1 N~~~. Cl 0.01 4-

bromoacetophenonen-butylacrylate 16 86.4 8640

4cwa—w-(3 0.1 4-


5

cop1"“Vo

0.01 4-bromoacetophenone


6r? JO r7 Me

D0.1 4-


7 LN '7 Me

O0.01

4-bromoacetophenone



8 r"N 1 1 )—Pd—ClY i .



9W~T~°| Me



109 ?

N 1 I Me

O



11

r^ 1

r f \ f1 Me

o



12f " >—Pd-CI

* Me

0.1 4-chlorobenzaldehyde


13| Me

b

0.1 4-chlorobenzaldehyde


14c e p

o —io— 0.1 4-chlorobenzaldehyde


Table 5.1: Results from testing of a number of preformed Pd complexes as catalysts in

the Heck reaction. (Reaction conditions: DMAc, 120°C, additive:

Tetrabutylammonium bromide. Each result is the average of at least two catalytic

runs.)

It can be seen from the table of results that the catalysts gave moderate conversions

with yields ranging from 8-88%. The time of 16 hours was an arbitrary time and

therefore the TONs stated do not necessarily represent the maximum turnovers, nor do

they indicate in anyway that the catalyst had ceased to operate. Indeed Danopoulos’s

similar system (entry 16 and 17, Table 5.2), was shown to have no loss in activity

with large increases in time, implying a high thermal stability under the reaction

conditions in a number of Heck catalytic runs11. It can be seen from the results that


the higher catalytic loading has an effect on the conversions by lowering the yield.

This is particularly evident for entries 1 to 3, with the 0.5, 0.1 and 0.01 mol% catalyst

loading giving 32, 50 and 86.4% conversions respectively. The reasons for this are

unclear though it may be possible that a decomposition pathway between two

molecules exist, the low catalyst loading having an effect of decreasing the frequency

of collisions. Both the dichloro- and methyl- chloro- complexes of the

octahydroacridine functionalised carbene were tested in the Heck reaction to evaluate

if the methyl group would have any significant advantages in this particular system.

Pd hydrocarbyl complexes of pyridine functionalised carbenes have been used in the

Heck reaction by both Danopoulos and Cavell in the coupling of BA and BAP (entries

17 and 18). It has been shown by Cavell that the methyl group has an activating

influence on catalyst formation, possibly by providing a facile route to the

catalytically active Pd(0) species through the insertion of the olefin and subsequent P-

hydride elimination4,9. This however doesn’t seem to have any significant effect in the

case of this system with TONs of 8640 and 8800 for the dichloro- (entry 3) and

methyl- chloro- (entry 5) systems respectively. The induction period was not

measured for this reaction and so the catalytic superiority for the methyl complexes

cannot be completely ruled out. Cavell also found the chloro complex of his ‘pincer’

complex (entry 20, Table 5.2) was consistently more active than the methyl derivative

(entry 19), though activity for the chloro- complex was found to decrease as the

catalytic cycle continued. On comparing entries 7 and 11, we can see the effect the

extra steric bulk of the methyl group in the 2 position of the quinoline function has on

the conversions. Complex 4.16 with the extra methyl group gave a 74% yield

compared to 40% without this group. When comparing the results for the different N-

substituents, i.e. the benzyl group and the methyl group, it can be seen that there is

very little difference with respect to conversions. On examination of the data in Table

5.2 for NHC-pyridine systems it is evident that the preformed catalysts used within

this work are comparable to those previously tested when factors such as reaction time

(16 hours versus 120 of entry 18), and reaction temperature (120°C versus 140°C of

entry 17) are taken into consideration.


Entry Catalyst Conditions/AdditivesTON

(TOF)

Coupling

Reagents

Run

Time

(Hours)

15

f V

cC 100°C >200

BAP &

BA-

16r O

r^N 1 [| X-Pd-BrMe

" T

140°C1400,000

(1,000)

Phi & MA

18

17r O

r^N I J V - p - B r

^ M e

■ " r

140°C200(11)

(100%)

BAP &

BA18

18r O

r—N I | [ V hj« ,

N Me

120°C610,000

(5,080)

BAP &

BA120

19[T > -P d -< J]

N 1 N1 1

120°C/Pr4NBr34,330

(1,720)

BAP &

BA20

20 ,-"NC y ~ ^ ji Cl ^

120°C and

n h 2n h 2.h 2o

40,160

(1,980)

BAP &

BA20


21 A

1 u 1 At At

1 4 0 ° C

1 4 , 2 9 0

( 7 9 3 )

BAP &

MA1 8

221 1At Cl Ar

1 4 0 ° C

1 4 , 2 9 0

( 7 9 3 )

BAP &

MA 1 8

Table 5.2. Selected literature examples of Pd NHC complexes in the Heck reaction.1 0(Adapted from the review by Crudden and Allen ).

Although none of the new complexes show particularly high activities for the

coupling of the aryl bromide, it was decided to evaluate their performance with an

activated aryl chloride. Entries 1 2 to 1 4 (Table 5 . 1 ) show the performance of

complexes 4.15, 4.16 and 4.20 with 4 - chlorobenzaldehyde. It can clearly be seen that

the complexes tested have very low activities with respect to the coupling 4 -

chlorobenzaldehyde and butyl acrylate with conversion from 8 - 1 4 % . There was no

evidence of Pd black in any of the catalytic runs indicating that the complexes were

stable throughout the length of the catalysis times.

The effect of rigidity of the chelating backbone in the Heck reaction can be compared

using McGuinness’s complex 5.3 (Figure 5 . 2 ) with those synthesised within this work

such as 4.14 and 4.20. Like those of this work, 5.3 forms a 6 membered chelate ring

and contains a second donor separated from the NHC by a -C - linker. The order of

rigidity should be 5.3< 4.20< 4.14. Table 5 . 3 shows some selected catalytic data from

McGuinness’s work with 5.39.


N

N

) F?d "CIN Me

5.3

Me

4.20

PdCII

Me

4.14

Figure 5.2: The effect of rigidity should increase from left to right due to the

nature of the linker between each donor.

EntryAmount

Mol%Aryl halide

Coupling

partnerTime Conversion TON

21 0.000164-

Bromoacetophenone

Butyl

acrylate48 63 393800

22 0.0214-

Chlorobenzaldehyde

Butyl

acrylate24 75 360

Table 5.3: Selected catalytic data from McGuinness’s work13. (120°C, DMAc)

The coupling of 4-bromoacetophenone and n-butyl acrylate with 5.3 and those

synthesised within this work are comparable with respect to conversions, indicating

that rigidity is not a key controlling factor for the Heck reaction. The high TON of

entry 21 reflects the catalyst loading in comparison to those of Table 5.1. If time

constraints were not an issue it would have been of interest to carry out the Heck

reaction using the same catalyst loading as in entry 21 for the Pd complexes

synthesised in Chapter Four, to see if the activities and hence TON’s would be similar.

The difference in reaction times must also be taken into consideration (48 hours

compared to 16) and so again it would be interesting to see if in fact it is the nature of


the ligand, or of the catalyst loading coupled with the reaction times, that provided

these similar conversions.

With respect to the coupling of 4-chlorobenzaldehyde it can be seen that

McGuinness’s complex out performed any of the complexes within this work, with a

satisfactory conversion of 75%, compared to our highest of 14% (entry 14). Again this

was achieved at lower catalyst loading (0.021mol% compared with 0.1mol%).

Further investigations with the complexes synthesised herein into the Heck reaction

should include the use of a range of aryl chlorides and bromides, a probe of the

induction time and a screening of the conditions with respect to reaction times and

catalyst loading, in order to fine tune the system for optimum results.

5.3 Conclusions

The majority of the complexes synthesised in Chapter Four were catalytically tested

in the Heck reaction in order to be benchmarked against the ever-growing number of

existing Pd functionalised carbene complexes. All of the complexes tested showed

low to satisfactory activities with respect to the coupling of 4-chlorobenzaldhyde and

4-bromoacetophenone with n-butyl acrylate, though are comparable to similar

systems when reaction conditions are taken into consideration. In terms of ligand

design it can be seen that the possession of a rigid, cyclic structure with delocalised

electrons has no obvious benefits in the Heck coupling of the aforementioned

substrates. In hindsight, it appears likely that Heck coupling was not the best catalytic

process to study when reviewing the catalytic performance of these new ligands. Due

to time constraints only the Heck reaction was investigated and so the successful

application of these complexes in other catalytic reactions cannot be ruled out.


5.4 Experimental

5.4.1 General comments

All reactions were performed under an atmosphere of dry nitrogen using standard

Schlenk techniques. All complexes used were synthesised according to the procedures

outlined in Chapter Four. Reagents were bought from commercial suppliers and used

without further purification. DMAc was dried over molecular sieves prior to use. GC-

MS analyses were carried out on a HP Agilent 5973, with an Agilent HPSMS column

(50m x 0.25mm).

5.4.2 Analysis Method

For all catalytic Heck reactions the yields of the coupled products were determined by

GC-MS based on the amount of aryl halide remaining after the allocated reaction time

using internal standard methods. GC-MS was used to identify the identity of the

desired product.

5.4.3 Procedure for the Heck Reaction

4-Bromoacetophenone (5mmol, 0.99g), sodium acetate (5.56mmol, 0.456g) and 20%

tetrabutylammoniumbromide (0.198g) was suspended in 5mL DMAc. n-butyl acrylate

(7mmol, lmL) was then added followed by the Pd catalyst in a DMAc solution. The

reaction mixture was then placed into a preheated oil bath at 120°C and allowed to stir

for 16 hours. After cooling, n-dodecane (5mmol, 1.13mL) was added to the mixture as

an internal standard. The mixture was then washed with water to remove inorganic

salts, and the solution analysed by GC-MS. GC-MS indicated only one product was

formed though selectivity was not measured.


5.5 References

1. I. Beletskaya, A. Cheprakov, Chem. Rev., 2000,100, 3009-3066.

2. W. Herrmann, M. Elison, J. Fischer, C. Kocher, G. Artus, Angew. Chem. Int.

Ed Eng., 1995,34(21), 2371-2374.

4. D. McGuinness, K. Cavell, M. Green, B. Skelton, A. White, J. Organomet.

Chem., 1998,165, 565.

5. A. Magill, D. McGuinness., K. Cavell, G. Britovsek, V. Gibson, A. White, D.

Williams, A. H. White, B. Skelton, J. Organomet. Chem., 2001, 546, 617-618.

6. W. Herrmann, V. Bohm, C. Gstottmayr, M. Grosche, C. Reisinger, T.

Weskamp, J. Organomet. Chem., 2001,617-618, 616-628.

7. W. Herrmann, C. Reisinger, M. Spiegler, J. Organomet. Chem., 1998, 557, 93.

8. K. Selvarkumar, A. Zapf, M. Beller, Organic Letters, 2002, 4, 3031-3033.

9. J. Schwartz, V.Bohm, M, Gardiner, M. Grosche, W. Herrmann, W. Hieringer,

G. Raudaschl-Sieber, Chem. Eur. J., 2000, 6, 1773.

10. T. Jeffery, J. Chem. Soc., Chem. Comm., 1984, 1287.

11. A. Tulloch, A. Danopoulos, R. Tooze, S. Cafferkey, S. Kleinhenz, M.

Hursthouse, Chem. Commun., 2000, 1247-1248.

12. C. M. Crudden, D. P. Allen, Coordination Chemistry Reviews, 2004, 248,

2247.

13. D. McGuinness, K. Cavell, Organometallics, 2000,19, 741-748.

123

Appendix

X- Ray crystallography

Crystal structure determination.

All single crystal X-ray data were collected on a Bruker Nonius Kappa CCD

diffractometer using graphite monochromated Mo-Ka radiation, equipped with an

Oxford Cryostream cooling apparatus.

l-methvl-3-(2-methvl-quinoHne)-imidazolium iodide (2.81

Empirical formula

Formula weight

Temperature

Wavelength

Crystal system

Space group

Unit cell dimensions

Volume

Z

Density (calculated)

Absorption coefficient

F(000)

Crystal size

Theta range for data collection

Index ranges

Reflections collected

Independent reflections

Completeness to theta = 26.37°

Absorption correction

Max. and min. transmission

C14H14IN3

351.18

150(2) K

0.71069 A Monoclinic

P 21/n

a = 10.523(5) A <x= 90.000(5)°.

b = 7.574(5) A p= 93.861(5)°.

c = 17.237(5) A y = 90.000(5)°.

1370.7(12) A3 4

1.702 Mg/m3

2.322 mm'1

688

0.23 x 0.23 x 0.15 mm3

3.57 to 26.37°.

-13<=h<=13, -9<=k<=9, -20<=1<=21

14387

2795 [R(int) = 0.1071]

99.7 %

Semi-empirical from equivalents

0.7220 and 0.6172

Refinement method

Data / restraints / parameters

Goodness-of-fit on F2

Final R indices [I>2sigma(I)]

R indices (all data)

Largest diff. peak and hole

Full-matrix least-squares on F2

2795/0 / 165

1.042

R1 = 0.0304, wR2 = 0.0642

R1 = 0.0398, wR2 = 0.0678

0.707 and -1.060 e .A 3

l-mesitvl-3-octahvdroacridinimidazolium chloride (2.14)

Empirical formula C27 H35 C18 N3 O

Formula weight 701.18

Temperature 150(2) K

Wavelength 0.71069 ACrystal system Monoclinic

Space group P 21/a

Unit cell dimensions a = 14.141(5) A a= 90.000(5)°.

b = 16.458(5) A p= 92.359(5)°.

c= 14.209(5) A y = 90.000(5)°.

Volume 3304.1(19) A3Z 4

Density (calculated) 1.410 Mg/m3

Absorption coefficient 0.708 m m 1

F(000) 1448

Crystal size 0.20 x 0.15 x 0.05 mm3

Theta range for data collection 3.13 to 27.46°.

Index ranges -18<=h<=18, -21<=k<=21, ■-14<=1<=18

Reflections collected 39529

Independent reflections 7557 [R(int) = 0.1688]

Completeness to theta = 27.46° 99.7 %

Absorption correction Semi-empirical from equivalents

Max. and min. transmission 0.9655 and 0.8714

Refinement method Full-matrix least-squares on F2

Data / restraints / parameters 7557/6/357

Goodness-of-fit on F2 1.018

Final R indices [I>2sigma(I)] R1 = 0.0903, wR2 = 0.2034

125

R indices (all data) R1 = 0.1578, wR2 = 0.2374

Largest diff. peak and hole 1.534 and -1.020 e.A'3

4,5-Bis(methvlimidazolium hexafluroDhosDhate)-2,2-dimethvl-13-dioxolan (2.23)

Empirical formula

Formula weight

Temperature

Wavelength

Crystal system

Space group


Volume

Z



F(000)

Crystal size


Index ranges






Refinement method





Absolute structure parameter


C15H24F12N4 02 P2

582.32

150(2) K

0.71073 AMonoclinic

P 2(1)a = 6.56670(10) A <x= 90°.

b = 14.9593(3) A p= 91.6244(11)°.

c= 12.1524(2) A y = 90°.

1193.29(4) A3 2

1.621 Mg/m3

0.296 m m 1

592

0.25 x 0.20 x 0.20 mm3

3.10 to 27.46°.

-8<=h<=8, -19<=k<=19, -15<=1<=15

20597

5354 [R(int) = 0.0790]

99.7 %


0.9431 and 0.9296


5354/ 1 /320

1.041

R1 =0.0419, wR2 = 0.1012

R1 = 0.0498, wR2 = 0.1064

-0.01(9)

0.249 and -0.320 e.A'3

126

Ag(I) complex of Octahydroacridine functionalised carbene (3.12).

Empirical formula

Formula weight

Temperature

Wavelength

Crystal system

Space group


Volume

Z



F(000)

Crystal size


Index ranges






Refinement method





C25 H29 Ag Cl N3

514.83

150(2) K


C 2/c

a = 15.5780(4) A b= 11.5620(3) A c = 25.8740(6) A 4508.60(19) A3 8

1.517 Mg/m3

1.030 mm'1

2112

0.45 x 0.33 x 0.33 mm3

2.95 to 27.51°.

-20<=h<=20, -14<=k<=14, -27<=1<=31

10294

4937 [R(int) = 0.0447]

95.0 %


0.7275 and 0.6544


4937 / 0 / 274

1.048

R1 = 0.0371, wR2 = 0.0809

R1 = 0.0517, wR2 = 0.0876

a= 90°.

p= 104.6560(10)°.

y = 90°.

Largest diff. peak and hole 0.690 and -0.828 e.A'3

IPd(Me)(l-methvl-3-quinoline-imidazolin-2-vlidene)Cll (4.13):

Empirical formula

Formula weight

Temperature

Wavelength

Crystal system

Space group


Volume

Z



F(000)

Crystal size


Index ranges






Refinement method



Pd-CII

Me

C14.50 H15 C12 N3 Pd

408.60

150(2) K

0.71073 AMonoclinic

P21/c

a = 15.8709(7) A a=

b= 13.6114(5) A p=

c = 15.9427(6) A y =3066.6(2) A3 8

1.770 Mg/m3

1.552 mm'1

1624

0.18 x 0.15 x 0.03 mm3

3.87 to 27.46°.

-20<=h<=20, -17<=k<=17, -20<=

27046

6909 [R(int) = 0.1531]

98.4 %


0.9549 and 0.7675


6909 / 0 / 374

1.048

90°.

117.075(2)°.

90°.

:=20




R1 =0.0591, wR2 = 0.1135

R1 = 0.0996, wR2 = 0.1276

0.799 and -0.758 e.A'3

[Pd(l-mesitvl-3-octahvdroacridinimidazolin-2-vlidene)CM (4.19):

Pd-CI

Empirical formula C26 H30 C14 N3 Pd

Formula weight 632.73

Temperature 150(2) K

Wavelength 0.71073 ACrystal system Monoclinic

Space group P 21/n

Unit cell dimensions a = 8.7030(2) A <x= 90°.

b = 17.6960(4) A p= 100.9000(10)°.

c = 17.3720(5) A y = 90°.

Volume 2627.16(11) A3Z 4

Density (calculated) 1.600 Mg/m3

Absorption coefficient 1.134 mm'1

F(000) 1284

Crystal size 0.40 x 0.38 x 0.20 mm3

Theta range for data collection 3.65 to 26.37°.

Index ranges -10<=h<=10, -22<=k<=22, -21<=1<=21

Reflections collected 20389

Independent reflections 5354 [R(int) = 0.0810]

Completeness to theta = 26.37° 99.6 %



Refinement method







0.8050 and 0.6597


5354/0 /310

1.058

R1 =0.0431, wR2 = 0.0905

R1 = 0.0674, wR2 = 0.0991

0.990 and -0.950 e.A 3

fPd(Me)(l-mesitvl-3-octahvdroacridinimidazolin-2-vlidene)C11 (4.20):

Pd-CIIMe

Empirical formula

Formula weight

Temperature

Wavelength

Crystal system

Space group


Volume

Z



F(000)

Crystal size


C27 H34 C13 N3 Pd

613.32

150(2) K


P 2(1 )/n

a = 9.6970(2) A b = 17.7540(5) A c= 15.5720(5) A 2662.75(13) A3 4

1.530 Mg/m3

1.019 m m 1

1256

0.38 x 0.20 x 0.20 mm3

3.12 to 27.45°.

a= 90°.

P= 96.6670(10)c

y = 90°.

130

Index ranges






Refinement method






-12<=h<=12, -22<=k<=23, -19<=1<=20

19490

6048 [R(int) = 0.0850]

99.3 %


0.8221 and 0.6981


6048/0/311

1.036

R1 = 0.0475, wR2 = 0.0964

R1 = 0.0775, wR2 = 0.1087

0.565 and -0.871 e.A'3

chelating carbene ligands and their metal complexeschelating carbene ligands and their metal...

Documents